talk-data.com talk-data.com

Filter by Source

Select conferences and events

People (125 results)

See all 125 →

Activities & events

Title & Speakers Event
Global AI EXPO 2025-12-10 · 18:00

Global AI Expo 2025 is a full-day, hands-on showcase of the biggest ideas in AI—built for people who learn by seeing and doing. Explore how AI is transforming industries, test emerging tools, and gain practical skills that prepare you for the future of work. You’ll roam across themed halls, join live demos and code-along workshops, network on your own terms, and—new this year—participate in a dedicated Global AI Job Fair in the final two hours.

When : Date : Dec 10, 2025 Time : From 1 pm - 5 pm (Your local Time) Format: Virtual - https://events.tao.ai/pod/analytics.club/9kpr6q1mqod1/source--me

What to Expect

  • Live Demonstrations – Watch AI in action: robotics, predictive analytics, conversational AI, autonomous systems, creative AI, and more.
  • Hands-On Workshops – Build, test, and refine AI models alongside developers, engineers, and researchers.
  • Networking Lounge – Connect with leaders, peers, and innovators in real-time.
  • AI Job Fair – Meet employers seeking AI talent and explore career opportunities across industries.

The Halls

  • Foundational AI – Explore the building blocks: core models, frameworks, and essential AI concepts.
  • Agentic AI – Dive into autonomous and reasoning agents that act, plan, and adapt in real-time.
  • Applied AI Exchange – See AI at work across industries: finance, health, energy, retail, and beyond.
  • Startup Launchpad – Discover bold startups, pitch sessions, and new AI ventures ready to disrupt.
  • Career & Talent Hub – Navigate AI career pathways, skills training, and job-matching opportunities.

Who Should Attend

  • Developers & Engineers pushing AI applications
  • Entrepreneurs building AI-enabled products
  • Students & Researchers applying AI concepts
  • Business Professionals operationalizing AI
  • Tech Enthusiasts eager to experiment
  • Professionals & Employers seeking AI talent

Why Join

  • Exclusive access to emerging AI tools and applications
  • Learn by doing—not just listening
  • Build your network with global AI professionals and innovators
  • Leave with actionable insights to fuel your projects or career

If you have any questions, please email us at : [email protected]

Tags AI jobs, #machine learning jobs, data science jobs, LLM engineer, MLOps engineer, AI recruiter, tech hiring, AI careers, ML internships, entry-level AI, senior AI roles, portfolio review, interview prep, resume tips, employer booths, remote AI roles, hiring fair, virtual job fair, career networking, talent matching, recruiters

Global AI EXPO
Global AI EXPO 2025-12-10 · 18:00

Global AI Expo 2025 is a full-day, hands-on showcase of the biggest ideas in AI—built for people who learn by seeing and doing. Explore how AI is transforming industries, test emerging tools, and gain practical skills that prepare you for the future of work. You’ll roam across themed halls, join live demos and code-along workshops, network on your own terms, and—new this year—participate in a dedicated Global AI Job Fair in the final two hours.

When : Date : Dec 10, 2025 Time : From 1 pm - 5 pm (Your local Time) Format: Virtual - https://events.tao.ai/pod/analytics.club/9kpr6q1mqod1/source--me

What to Expect

  • Live Demonstrations – Watch AI in action: robotics, predictive analytics, conversational AI, autonomous systems, creative AI, and more.
  • Hands-On Workshops – Build, test, and refine AI models alongside developers, engineers, and researchers.
  • Networking Lounge – Connect with leaders, peers, and innovators in real-time.
  • AI Job Fair – Meet employers seeking AI talent and explore career opportunities across industries.

The Halls

  • Foundational AI – Explore the building blocks: core models, frameworks, and essential AI concepts.
  • Agentic AI – Dive into autonomous and reasoning agents that act, plan, and adapt in real-time.
  • Applied AI Exchange – See AI at work across industries: finance, health, energy, retail, and beyond.
  • Startup Launchpad – Discover bold startups, pitch sessions, and new AI ventures ready to disrupt.
  • Career & Talent Hub – Navigate AI career pathways, skills training, and job-matching opportunities.

Who Should Attend

  • Developers & Engineers pushing AI applications
  • Entrepreneurs building AI-enabled products
  • Students & Researchers applying AI concepts
  • Business Professionals operationalizing AI
  • Tech Enthusiasts eager to experiment
  • Professionals & Employers seeking AI talent

Why Join

  • Exclusive access to emerging AI tools and applications
  • Learn by doing—not just listening
  • Build your network with global AI professionals and innovators
  • Leave with actionable insights to fuel your projects or career

If you have any questions, please email us at : [email protected]

Tags AI jobs, #machine learning jobs, data science jobs, LLM engineer, MLOps engineer, AI recruiter, tech hiring, AI careers, ML internships, entry-level AI, senior AI roles, portfolio review, interview prep, resume tips, employer booths, remote AI roles, hiring fair, virtual job fair, career networking, talent matching, recruiters

Global AI EXPO
Global AI EXPO 2025-12-10 · 13:00

Global AI Expo 2025 is a full-day, hands-on showcase of the biggest ideas in AI—built for people who learn by seeing and doing. Explore how AI is transforming industries, test emerging tools, and gain practical skills that prepare you for the future of work. You’ll roam across themed halls, join live demos and code-along workshops, network on your own terms, and—new this year—participate in a dedicated Global AI Job Fair in the final two hours.

When : Date : Dec 10, 2025 Time : From 1 pm - 5 pm (Your local Time) Format: Virtual - https://events.tao.ai/pod/analytics.club/9kpr6q1mqod1/source--me

What to Expect

  • Live Demonstrations – Watch AI in action: robotics, predictive analytics, conversational AI, autonomous systems, creative AI, and more.
  • Hands-On Workshops – Build, test, and refine AI models alongside developers, engineers, and researchers.
  • Networking Lounge – Connect with leaders, peers, and innovators in real-time.
  • AI Job Fair – Meet employers seeking AI talent and explore career opportunities across industries.

The Halls

  • Foundational AI – Explore the building blocks: core models, frameworks, and essential AI concepts.
  • Agentic AI – Dive into autonomous and reasoning agents that act, plan, and adapt in real-time.
  • Applied AI Exchange – See AI at work across industries: finance, health, energy, retail, and beyond.
  • Startup Launchpad – Discover bold startups, pitch sessions, and new AI ventures ready to disrupt.
  • Career & Talent Hub – Navigate AI career pathways, skills training, and job-matching opportunities.

Who Should Attend

  • Developers & Engineers pushing AI applications
  • Entrepreneurs building AI-enabled products
  • Students & Researchers applying AI concepts
  • Business Professionals operationalizing AI
  • Tech Enthusiasts eager to experiment
  • Professionals & Employers seeking AI talent

Why Join

  • Exclusive access to emerging AI tools and applications
  • Learn by doing—not just listening
  • Build your network with global AI professionals and innovators
  • Leave with actionable insights to fuel your projects or career

If you have any questions, please email us at : [email protected]

Tags AI jobs, #machine learning jobs, data science jobs, LLM engineer, MLOps engineer, AI recruiter, tech hiring, AI careers, ML internships, entry-level AI, senior AI roles, portfolio review, interview prep, resume tips, employer booths, remote AI roles, hiring fair, virtual job fair, career networking, talent matching, recruiters

Global AI EXPO
Global AI EXPO 2025-12-10 · 12:00

Global AI Expo 2025 is a full-day, hands-on showcase of the biggest ideas in AI—built for people who learn by seeing and doing. Explore how AI is transforming industries, test emerging tools, and gain practical skills that prepare you for the future of work. You’ll roam across themed halls, join live demos and code-along workshops, network on your own terms, and—new this year—participate in a dedicated Global AI Job Fair in the final two hours.

When : Date : Dec 10, 2025 Time : From 1 pm - 5 pm (Your local Time) Format: Virtual - https://events.tao.ai/pod/analytics.club/9kpr6q1mqod1/source--me

What to Expect

  • Live Demonstrations – Watch AI in action: robotics, predictive analytics, conversational AI, autonomous systems, creative AI, and more.
  • Hands-On Workshops – Build, test, and refine AI models alongside developers, engineers, and researchers.
  • Networking Lounge – Connect with leaders, peers, and innovators in real-time.
  • AI Job Fair – Meet employers seeking AI talent and explore career opportunities across industries.

The Halls

  • Foundational AI – Explore the building blocks: core models, frameworks, and essential AI concepts.
  • Agentic AI – Dive into autonomous and reasoning agents that act, plan, and adapt in real-time.
  • Applied AI Exchange – See AI at work across industries: finance, health, energy, retail, and beyond.
  • Startup Launchpad – Discover bold startups, pitch sessions, and new AI ventures ready to disrupt.
  • Career & Talent Hub – Navigate AI career pathways, skills training, and job-matching opportunities.

Who Should Attend

  • Developers & Engineers pushing AI applications
  • Entrepreneurs building AI-enabled products
  • Students & Researchers applying AI concepts
  • Business Professionals operationalizing AI
  • Tech Enthusiasts eager to experiment
  • Professionals & Employers seeking AI talent

Why Join

  • Exclusive access to emerging AI tools and applications
  • Learn by doing—not just listening
  • Build your network with global AI professionals and innovators
  • Leave with actionable insights to fuel your projects or career

If you have any questions, please email us at : [email protected]

Tags AI jobs, #machine learning jobs, data science jobs, LLM engineer, MLOps engineer, AI recruiter, tech hiring, AI careers, ML internships, entry-level AI, senior AI roles, portfolio review, interview prep, resume tips, employer booths, remote AI roles, hiring fair, virtual job fair, career networking, talent matching, recruiters

Global AI EXPO
Global AI EXPO 2025-12-10 · 12:00

Global AI Expo 2025 is a full-day, hands-on showcase of the biggest ideas in AI—built for people who learn by seeing and doing. Explore how AI is transforming industries, test emerging tools, and gain practical skills that prepare you for the future of work. You’ll roam across themed halls, join live demos and code-along workshops, network on your own terms, and—new this year—participate in a dedicated Global AI Job Fair in the final two hours.

When : Date : Dec 10, 2025 Time : From 1 pm - 5 pm (Your local Time) Format: Virtual - https://events.tao.ai/pod/analytics.club/9kpr6q1mqod1/source--me

What to Expect

  • Live Demonstrations – Watch AI in action: robotics, predictive analytics, conversational AI, autonomous systems, creative AI, and more.
  • Hands-On Workshops – Build, test, and refine AI models alongside developers, engineers, and researchers.
  • Networking Lounge – Connect with leaders, peers, and innovators in real-time.
  • AI Job Fair – Meet employers seeking AI talent and explore career opportunities across industries.

The Halls

  • Foundational AI – Explore the building blocks: core models, frameworks, and essential AI concepts.
  • Agentic AI – Dive into autonomous and reasoning agents that act, plan, and adapt in real-time.
  • Applied AI Exchange – See AI at work across industries: finance, health, energy, retail, and beyond.
  • Startup Launchpad – Discover bold startups, pitch sessions, and new AI ventures ready to disrupt.
  • Career & Talent Hub – Navigate AI career pathways, skills training, and job-matching opportunities.

Who Should Attend

  • Developers & Engineers pushing AI applications
  • Entrepreneurs building AI-enabled products
  • Students & Researchers applying AI concepts
  • Business Professionals operationalizing AI
  • Tech Enthusiasts eager to experiment
  • Professionals & Employers seeking AI talent

Why Join

  • Exclusive access to emerging AI tools and applications
  • Learn by doing—not just listening
  • Build your network with global AI professionals and innovators
  • Leave with actionable insights to fuel your projects or career

If you have any questions, please email us at : [email protected]

Tags AI jobs, #machine learning jobs, data science jobs, LLM engineer, MLOps engineer, AI recruiter, tech hiring, AI careers, ML internships, entry-level AI, senior AI roles, portfolio review, interview prep, resume tips, employer booths, remote AI roles, hiring fair, virtual job fair, career networking, talent matching, recruiters

Global AI EXPO
Global AI EXPO 2025-12-10 · 12:00

Global AI Expo 2025 is a full-day, hands-on showcase of the biggest ideas in AI—built for people who learn by seeing and doing. Explore how AI is transforming industries, test emerging tools, and gain practical skills that prepare you for the future of work. You’ll roam across themed halls, join live demos and code-along workshops, network on your own terms, and—new this year—participate in a dedicated Global AI Job Fair in the final two hours.

When : Date : Dec 10, 2025 Time : From 1 pm - 5 pm (Your local Time) Format: Virtual - https://events.tao.ai/pod/analytics.club/9kpr6q1mqod1/source--me

What to Expect

  • Live Demonstrations – Watch AI in action: robotics, predictive analytics, conversational AI, autonomous systems, creative AI, and more.
  • Hands-On Workshops – Build, test, and refine AI models alongside developers, engineers, and researchers.
  • Networking Lounge – Connect with leaders, peers, and innovators in real-time.
  • AI Job Fair – Meet employers seeking AI talent and explore career opportunities across industries.

The Halls

  • Foundational AI – Explore the building blocks: core models, frameworks, and essential AI concepts.
  • Agentic AI – Dive into autonomous and reasoning agents that act, plan, and adapt in real-time.
  • Applied AI Exchange – See AI at work across industries: finance, health, energy, retail, and beyond.
  • Startup Launchpad – Discover bold startups, pitch sessions, and new AI ventures ready to disrupt.
  • Career & Talent Hub – Navigate AI career pathways, skills training, and job-matching opportunities.

Who Should Attend

  • Developers & Engineers pushing AI applications
  • Entrepreneurs building AI-enabled products
  • Students & Researchers applying AI concepts
  • Business Professionals operationalizing AI
  • Tech Enthusiasts eager to experiment
  • Professionals & Employers seeking AI talent

Why Join

  • Exclusive access to emerging AI tools and applications
  • Learn by doing—not just listening
  • Build your network with global AI professionals and innovators
  • Leave with actionable insights to fuel your projects or career

If you have any questions, please email us at : [email protected]

Tags AI jobs, #machine learning jobs, data science jobs, LLM engineer, MLOps engineer, AI recruiter, tech hiring, AI careers, ML internships, entry-level AI, senior AI roles, portfolio review, interview prep, resume tips, employer booths, remote AI roles, hiring fair, virtual job fair, career networking, talent matching, recruiters

Global AI EXPO

This month we’re delighted to be joined by Tesco’s Head of Data and we’ll also hear about the latest in AI Governance on Databricks from the Advancing Analytics team!

17:30 - 18:00: Arrival & Networking 18:00 - 18:10: Opening Remarks & Introductions

18:10 - 18:40: Building Sustainable Data Products with AI Readiness at Scale for Cyber - Varun S Gangoor - Head of Data at Tesco In this session, we’ll explore how sustainable data product design can enable AI readiness and drive innovation in cybersecurity. We’ll look at the key principles behind building scalable, secure and reusable data products that support cyber analytics, machine learning and GenAI use cases at scale. We’ll also discuss practical approaches to data governance, automation and platform engineering that help cyber teams turn data into actionable intelligence and strengthen overall cyber resilience.

18.40-19.10 Governing AI In Databricks: Building Trust And Compliance At Scale - Gavi Regunath, Advancing Analytics CAIO and Databricks MVP, joined by Terry McCann, Advancing Analytics CEO AI governance isn’t optional - it’s foundational. With 45% of organisations concerned about data accuracy and bias, and 40% worried about privacy, the need for robust oversight is clear. From model safety to compliance tooling, we’ll show how to embed trust into your AI systems, without slowing innovation.

19:00 onwards: Pizza, Drinks & Networking Enjoy some delicious pizza and beverages while networking with peers.

Join us for a fantastic evening of learning and networking at the London Databricks meetup!

November 2025 - Databricks Meetup

Join the virtual Meetup to hear talks from experts on cutting-edge topics across AI, ML, and computer vision.

Date, Time and Location

Oct 30, 2025 9 AM Pacific Online. Register for the Zoom!

The Agent Factory: Building a Platform for Enterprise-Wide AI Automation

In this talk we will explore what it takes to build an enterprise-ready AI automation platform at scale. The topics covered will include:

  • The Scale Challenge: E-commerce environments expose the limitations of single-point AI solutions, which create fragmented ecosystems lacking cohesion and efficient resource sharing across complex, knowledge-based work.
  • Root Cause Analysis Success: Flipkart’s initial AI agent transformed business analysis from days-long investigations to near-instantaneous insights, proving the concept while revealing broader platform opportunities.
  • Platform Strategy Evolution: Success across Engineering (SDLC, SRE), Operations, and Commerce teams necessitated a unified, multi-tenant platform serving diverse use cases with consistency and operational efficiency.
  • Architectural Foundation: Leveraging framework-agnostic design principles we were able to emphasize modularity, which enabled teams to leverage different AI models while maintaining consistent interfaces and scalable infrastructure.
  • The “Agent Garden” Vision: Flipkart’s roadmap envisions an internal ecosystem where teams discover, deploy, and contribute AI agents, providing a practical blueprint for scalable AI agent infrastructure development.

About the Speaker

Virender Bhargav at Flipkart is a seasoned engineering leader whose expertise spans business technology integration, enterprise applications, system design/architecture, and building highly scalable systems. With a deep understanding of technology, he has spearheaded teams, modernized technology landscapes, and managed core platform layers and strategic products. With extensive experience driving innovation at companies like Paytm and Flipkart, his contributions have left a lasting impact on the industry.

Scaling Generative Models at Scale with Ray and PyTorch

Generative image models like Stable Diffusion have opened up exciting possibilities for personalization, creativity, and scalable deployment. However, fine-tuning them in production‐grade settings poses challenges: managing compute, hyperparameters, model size, data, and distributed coordination are nontrivial.

In this talk, we’ll dive deep into learning how to fine-tune Stable Diffusion models using Ray Train (with HuggingFace Diffusers), including approaches like DreamBooth and LoRA. We’ll cover what works (and what doesn’t) in scaling out training jobs, handling large data, optimizing for GPU memory and speed, and validating outputs. Attendees will come away with practical insights and patterns they can use to fine-tune generative models in their own work.

About the Speaker

Suman Debnath is a Technical Lead (ML) at Anyscale, where he focuses on distributed training, fine-tuning, and inference optimization at scale on the cloud. His work centers around building and optimizing end-to-end machine learning workflows powered by distributed computing framework like Ray, enabling scalable and efficient ML systems. Suman’s expertise spans Natural Language Processing (NLP), Large Language Models (LLMs), and Retrieval-Augmented Generation (RAG). Earlier in his career, he developed performance benchmarking and monitoring tools for distributed storage systems. Beyond engineering, Suman is an active community contributor, having spoken at over 100 global conferences and events, including PyCon, PyData, ODSC, AIE and numerous meetups worldwide.

Privacy-preserving in Computer Vision through Optics Learning

Cameras are now ubiquitous, powering computer vision systems that assist us in everyday tasks and critical settings such as operating rooms. Yet, their widespread use raises serious privacy concerns: traditional cameras are designed to capture high-resolution images, making it easy to identify sensitive attributes such as faces, nudity, or personal objects. Once acquired, such data can be misused if accessed by adversaries. Existing software-based privacy mechanisms, such as blurring or pixelation, often degrade task performance and leave vulnerabilities in the processing pipeline.

In this talk, we explore an alternative question: how can we preserve privacy before or during image acquisition? By revisiting the image formation model, we show how camera optics themselves can be learned and optimized to acquire images that are unintelligible to humans yet remain useful for downstream vision tasks like action recognition. We will discuss recent approaches to learning camera lenses that intentionally produce privacy-preserving images, blurry and unrecognizable to the human eye, but still effective for machine perception. This paradigm shift opens the door to a new generation of cameras that embed privacy directly into their hardware design.

About the Speaker

Carlos Hinojosa is a Postdoctoral researcher at King Abdullah University of Science and Technology (KAUST) working with Prof. Bernard Ghanem. His research interests span Computer Vision, Machine Learning, AI Safety, and AI for Science. He focuses on developing safe, accurate, and efficient vision systems and machine-learning models that can reliably perceive, understand, and act on information, while ensuring robustness, protecting privacy, and aligning with societal values.

It's a (Blind) Match! Towards Vision-Language Correspondence without Parallel Data

Can we match vision and language embeddings without any supervision? According to the platonic representation hypothesis, as model and dataset scales increase, distances between corresponding representations are becoming similar in both embedding spaces. Our study demonstrates that pairwise distances are often sufficient to enable unsupervised matching, allowing vision-language correspondences to be discovered without any parallel data.

About the Speaker

Dominik Schnaus is a third-year Ph.D. student in the Computer Vision Group at the Technical University of Munich (TUM), supervised by Daniel Cremers. His research centers on multimodal and self-supervised learning with a special emphasis on understanding similarities across embedding spaces of different modalities.

Oct 30 - AI, ML and Computer Vision Meetup

Join the virtual Meetup to hear talks from experts on cutting-edge topics across AI, ML, and computer vision.

Date, Time and Location

Oct 30, 2025 9 AM Pacific Online. Register for the Zoom!

The Agent Factory: Building a Platform for Enterprise-Wide AI Automation

In this talk we will explore what it takes to build an enterprise-ready AI automation platform at scale. The topics covered will include:

  • The Scale Challenge: E-commerce environments expose the limitations of single-point AI solutions, which create fragmented ecosystems lacking cohesion and efficient resource sharing across complex, knowledge-based work.
  • Root Cause Analysis Success: Flipkart’s initial AI agent transformed business analysis from days-long investigations to near-instantaneous insights, proving the concept while revealing broader platform opportunities.
  • Platform Strategy Evolution: Success across Engineering (SDLC, SRE), Operations, and Commerce teams necessitated a unified, multi-tenant platform serving diverse use cases with consistency and operational efficiency.
  • Architectural Foundation: Leveraging framework-agnostic design principles we were able to emphasize modularity, which enabled teams to leverage different AI models while maintaining consistent interfaces and scalable infrastructure.
  • The “Agent Garden” Vision: Flipkart’s roadmap envisions an internal ecosystem where teams discover, deploy, and contribute AI agents, providing a practical blueprint for scalable AI agent infrastructure development.

About the Speaker

Virender Bhargav at Flipkart is a seasoned engineering leader whose expertise spans business technology integration, enterprise applications, system design/architecture, and building highly scalable systems. With a deep understanding of technology, he has spearheaded teams, modernized technology landscapes, and managed core platform layers and strategic products. With extensive experience driving innovation at companies like Paytm and Flipkart, his contributions have left a lasting impact on the industry.

Scaling Generative Models at Scale with Ray and PyTorch

Generative image models like Stable Diffusion have opened up exciting possibilities for personalization, creativity, and scalable deployment. However, fine-tuning them in production‐grade settings poses challenges: managing compute, hyperparameters, model size, data, and distributed coordination are nontrivial.

In this talk, we’ll dive deep into learning how to fine-tune Stable Diffusion models using Ray Train (with HuggingFace Diffusers), including approaches like DreamBooth and LoRA. We’ll cover what works (and what doesn’t) in scaling out training jobs, handling large data, optimizing for GPU memory and speed, and validating outputs. Attendees will come away with practical insights and patterns they can use to fine-tune generative models in their own work.

About the Speaker

Suman Debnath is a Technical Lead (ML) at Anyscale, where he focuses on distributed training, fine-tuning, and inference optimization at scale on the cloud. His work centers around building and optimizing end-to-end machine learning workflows powered by distributed computing framework like Ray, enabling scalable and efficient ML systems. Suman’s expertise spans Natural Language Processing (NLP), Large Language Models (LLMs), and Retrieval-Augmented Generation (RAG). Earlier in his career, he developed performance benchmarking and monitoring tools for distributed storage systems. Beyond engineering, Suman is an active community contributor, having spoken at over 100 global conferences and events, including PyCon, PyData, ODSC, AIE and numerous meetups worldwide.

Privacy-preserving in Computer Vision through Optics Learning

Cameras are now ubiquitous, powering computer vision systems that assist us in everyday tasks and critical settings such as operating rooms. Yet, their widespread use raises serious privacy concerns: traditional cameras are designed to capture high-resolution images, making it easy to identify sensitive attributes such as faces, nudity, or personal objects. Once acquired, such data can be misused if accessed by adversaries. Existing software-based privacy mechanisms, such as blurring or pixelation, often degrade task performance and leave vulnerabilities in the processing pipeline.

In this talk, we explore an alternative question: how can we preserve privacy before or during image acquisition? By revisiting the image formation model, we show how camera optics themselves can be learned and optimized to acquire images that are unintelligible to humans yet remain useful for downstream vision tasks like action recognition. We will discuss recent approaches to learning camera lenses that intentionally produce privacy-preserving images, blurry and unrecognizable to the human eye, but still effective for machine perception. This paradigm shift opens the door to a new generation of cameras that embed privacy directly into their hardware design.

About the Speaker

Carlos Hinojosa is a Postdoctoral researcher at King Abdullah University of Science and Technology (KAUST) working with Prof. Bernard Ghanem. His research interests span Computer Vision, Machine Learning, AI Safety, and AI for Science. He focuses on developing safe, accurate, and efficient vision systems and machine-learning models that can reliably perceive, understand, and act on information, while ensuring robustness, protecting privacy, and aligning with societal values.

It's a (Blind) Match! Towards Vision-Language Correspondence without Parallel Data

Can we match vision and language embeddings without any supervision? According to the platonic representation hypothesis, as model and dataset scales increase, distances between corresponding representations are becoming similar in both embedding spaces. Our study demonstrates that pairwise distances are often sufficient to enable unsupervised matching, allowing vision-language correspondences to be discovered without any parallel data.

About the Speaker

Dominik Schnaus is a third-year Ph.D. student in the Computer Vision Group at the Technical University of Munich (TUM), supervised by Daniel Cremers. His research centers on multimodal and self-supervised learning with a special emphasis on understanding similarities across embedding spaces of different modalities.

Oct 30 - AI, ML and Computer Vision Meetup

Join the virtual Meetup to hear talks from experts on cutting-edge topics across AI, ML, and computer vision.

Date, Time and Location

Oct 30, 2025 9 AM Pacific Online. Register for the Zoom!

The Agent Factory: Building a Platform for Enterprise-Wide AI Automation

In this talk we will explore what it takes to build an enterprise-ready AI automation platform at scale. The topics covered will include:

  • The Scale Challenge: E-commerce environments expose the limitations of single-point AI solutions, which create fragmented ecosystems lacking cohesion and efficient resource sharing across complex, knowledge-based work.
  • Root Cause Analysis Success: Flipkart’s initial AI agent transformed business analysis from days-long investigations to near-instantaneous insights, proving the concept while revealing broader platform opportunities.
  • Platform Strategy Evolution: Success across Engineering (SDLC, SRE), Operations, and Commerce teams necessitated a unified, multi-tenant platform serving diverse use cases with consistency and operational efficiency.
  • Architectural Foundation: Leveraging framework-agnostic design principles we were able to emphasize modularity, which enabled teams to leverage different AI models while maintaining consistent interfaces and scalable infrastructure.
  • The “Agent Garden” Vision: Flipkart’s roadmap envisions an internal ecosystem where teams discover, deploy, and contribute AI agents, providing a practical blueprint for scalable AI agent infrastructure development.

About the Speaker

Virender Bhargav at Flipkart is a seasoned engineering leader whose expertise spans business technology integration, enterprise applications, system design/architecture, and building highly scalable systems. With a deep understanding of technology, he has spearheaded teams, modernized technology landscapes, and managed core platform layers and strategic products. With extensive experience driving innovation at companies like Paytm and Flipkart, his contributions have left a lasting impact on the industry.

Scaling Generative Models at Scale with Ray and PyTorch

Generative image models like Stable Diffusion have opened up exciting possibilities for personalization, creativity, and scalable deployment. However, fine-tuning them in production‐grade settings poses challenges: managing compute, hyperparameters, model size, data, and distributed coordination are nontrivial.

In this talk, we’ll dive deep into learning how to fine-tune Stable Diffusion models using Ray Train (with HuggingFace Diffusers), including approaches like DreamBooth and LoRA. We’ll cover what works (and what doesn’t) in scaling out training jobs, handling large data, optimizing for GPU memory and speed, and validating outputs. Attendees will come away with practical insights and patterns they can use to fine-tune generative models in their own work.

About the Speaker

Suman Debnath is a Technical Lead (ML) at Anyscale, where he focuses on distributed training, fine-tuning, and inference optimization at scale on the cloud. His work centers around building and optimizing end-to-end machine learning workflows powered by distributed computing framework like Ray, enabling scalable and efficient ML systems. Suman’s expertise spans Natural Language Processing (NLP), Large Language Models (LLMs), and Retrieval-Augmented Generation (RAG). Earlier in his career, he developed performance benchmarking and monitoring tools for distributed storage systems. Beyond engineering, Suman is an active community contributor, having spoken at over 100 global conferences and events, including PyCon, PyData, ODSC, AIE and numerous meetups worldwide.

Privacy-preserving in Computer Vision through Optics Learning

Cameras are now ubiquitous, powering computer vision systems that assist us in everyday tasks and critical settings such as operating rooms. Yet, their widespread use raises serious privacy concerns: traditional cameras are designed to capture high-resolution images, making it easy to identify sensitive attributes such as faces, nudity, or personal objects. Once acquired, such data can be misused if accessed by adversaries. Existing software-based privacy mechanisms, such as blurring or pixelation, often degrade task performance and leave vulnerabilities in the processing pipeline.

In this talk, we explore an alternative question: how can we preserve privacy before or during image acquisition? By revisiting the image formation model, we show how camera optics themselves can be learned and optimized to acquire images that are unintelligible to humans yet remain useful for downstream vision tasks like action recognition. We will discuss recent approaches to learning camera lenses that intentionally produce privacy-preserving images, blurry and unrecognizable to the human eye, but still effective for machine perception. This paradigm shift opens the door to a new generation of cameras that embed privacy directly into their hardware design.

About the Speaker

Carlos Hinojosa is a Postdoctoral researcher at King Abdullah University of Science and Technology (KAUST) working with Prof. Bernard Ghanem. His research interests span Computer Vision, Machine Learning, AI Safety, and AI for Science. He focuses on developing safe, accurate, and efficient vision systems and machine-learning models that can reliably perceive, understand, and act on information, while ensuring robustness, protecting privacy, and aligning with societal values.

It's a (Blind) Match! Towards Vision-Language Correspondence without Parallel Data

Can we match vision and language embeddings without any supervision? According to the platonic representation hypothesis, as model and dataset scales increase, distances between corresponding representations are becoming similar in both embedding spaces. Our study demonstrates that pairwise distances are often sufficient to enable unsupervised matching, allowing vision-language correspondences to be discovered without any parallel data.

About the Speaker

Dominik Schnaus is a third-year Ph.D. student in the Computer Vision Group at the Technical University of Munich (TUM), supervised by Daniel Cremers. His research centers on multimodal and self-supervised learning with a special emphasis on understanding similarities across embedding spaces of different modalities.

Oct 30 - AI, ML and Computer Vision Meetup

Join the virtual Meetup to hear talks from experts on cutting-edge topics across AI, ML, and computer vision.

Date, Time and Location

Oct 30, 2025 9 AM Pacific Online. Register for the Zoom!

The Agent Factory: Building a Platform for Enterprise-Wide AI Automation

In this talk we will explore what it takes to build an enterprise-ready AI automation platform at scale. The topics covered will include:

  • The Scale Challenge: E-commerce environments expose the limitations of single-point AI solutions, which create fragmented ecosystems lacking cohesion and efficient resource sharing across complex, knowledge-based work.
  • Root Cause Analysis Success: Flipkart’s initial AI agent transformed business analysis from days-long investigations to near-instantaneous insights, proving the concept while revealing broader platform opportunities.
  • Platform Strategy Evolution: Success across Engineering (SDLC, SRE), Operations, and Commerce teams necessitated a unified, multi-tenant platform serving diverse use cases with consistency and operational efficiency.
  • Architectural Foundation: Leveraging framework-agnostic design principles we were able to emphasize modularity, which enabled teams to leverage different AI models while maintaining consistent interfaces and scalable infrastructure.
  • The “Agent Garden” Vision: Flipkart’s roadmap envisions an internal ecosystem where teams discover, deploy, and contribute AI agents, providing a practical blueprint for scalable AI agent infrastructure development.

About the Speaker

Virender Bhargav at Flipkart is a seasoned engineering leader whose expertise spans business technology integration, enterprise applications, system design/architecture, and building highly scalable systems. With a deep understanding of technology, he has spearheaded teams, modernized technology landscapes, and managed core platform layers and strategic products. With extensive experience driving innovation at companies like Paytm and Flipkart, his contributions have left a lasting impact on the industry.

Scaling Generative Models at Scale with Ray and PyTorch

Generative image models like Stable Diffusion have opened up exciting possibilities for personalization, creativity, and scalable deployment. However, fine-tuning them in production‐grade settings poses challenges: managing compute, hyperparameters, model size, data, and distributed coordination are nontrivial.

In this talk, we’ll dive deep into learning how to fine-tune Stable Diffusion models using Ray Train (with HuggingFace Diffusers), including approaches like DreamBooth and LoRA. We’ll cover what works (and what doesn’t) in scaling out training jobs, handling large data, optimizing for GPU memory and speed, and validating outputs. Attendees will come away with practical insights and patterns they can use to fine-tune generative models in their own work.

About the Speaker

Suman Debnath is a Technical Lead (ML) at Anyscale, where he focuses on distributed training, fine-tuning, and inference optimization at scale on the cloud. His work centers around building and optimizing end-to-end machine learning workflows powered by distributed computing framework like Ray, enabling scalable and efficient ML systems. Suman’s expertise spans Natural Language Processing (NLP), Large Language Models (LLMs), and Retrieval-Augmented Generation (RAG). Earlier in his career, he developed performance benchmarking and monitoring tools for distributed storage systems. Beyond engineering, Suman is an active community contributor, having spoken at over 100 global conferences and events, including PyCon, PyData, ODSC, AIE and numerous meetups worldwide.

Privacy-preserving in Computer Vision through Optics Learning

Cameras are now ubiquitous, powering computer vision systems that assist us in everyday tasks and critical settings such as operating rooms. Yet, their widespread use raises serious privacy concerns: traditional cameras are designed to capture high-resolution images, making it easy to identify sensitive attributes such as faces, nudity, or personal objects. Once acquired, such data can be misused if accessed by adversaries. Existing software-based privacy mechanisms, such as blurring or pixelation, often degrade task performance and leave vulnerabilities in the processing pipeline.

In this talk, we explore an alternative question: how can we preserve privacy before or during image acquisition? By revisiting the image formation model, we show how camera optics themselves can be learned and optimized to acquire images that are unintelligible to humans yet remain useful for downstream vision tasks like action recognition. We will discuss recent approaches to learning camera lenses that intentionally produce privacy-preserving images, blurry and unrecognizable to the human eye, but still effective for machine perception. This paradigm shift opens the door to a new generation of cameras that embed privacy directly into their hardware design.

About the Speaker

Carlos Hinojosa is a Postdoctoral researcher at King Abdullah University of Science and Technology (KAUST) working with Prof. Bernard Ghanem. His research interests span Computer Vision, Machine Learning, AI Safety, and AI for Science. He focuses on developing safe, accurate, and efficient vision systems and machine-learning models that can reliably perceive, understand, and act on information, while ensuring robustness, protecting privacy, and aligning with societal values.

It's a (Blind) Match! Towards Vision-Language Correspondence without Parallel Data

Can we match vision and language embeddings without any supervision? According to the platonic representation hypothesis, as model and dataset scales increase, distances between corresponding representations are becoming similar in both embedding spaces. Our study demonstrates that pairwise distances are often sufficient to enable unsupervised matching, allowing vision-language correspondences to be discovered without any parallel data.

About the Speaker

Dominik Schnaus is a third-year Ph.D. student in the Computer Vision Group at the Technical University of Munich (TUM), supervised by Daniel Cremers. His research centers on multimodal and self-supervised learning with a special emphasis on understanding similarities across embedding spaces of different modalities.

Oct 30 - AI, ML and Computer Vision Meetup

Join the virtual Meetup to hear talks from experts on cutting-edge topics across AI, ML, and computer vision.

Date, Time and Location

Oct 30, 2025 9 AM Pacific Online. Register for the Zoom!

The Agent Factory: Building a Platform for Enterprise-Wide AI Automation

In this talk we will explore what it takes to build an enterprise-ready AI automation platform at scale. The topics covered will include:

  • The Scale Challenge: E-commerce environments expose the limitations of single-point AI solutions, which create fragmented ecosystems lacking cohesion and efficient resource sharing across complex, knowledge-based work.
  • Root Cause Analysis Success: Flipkart’s initial AI agent transformed business analysis from days-long investigations to near-instantaneous insights, proving the concept while revealing broader platform opportunities.
  • Platform Strategy Evolution: Success across Engineering (SDLC, SRE), Operations, and Commerce teams necessitated a unified, multi-tenant platform serving diverse use cases with consistency and operational efficiency.
  • Architectural Foundation: Leveraging framework-agnostic design principles we were able to emphasize modularity, which enabled teams to leverage different AI models while maintaining consistent interfaces and scalable infrastructure.
  • The “Agent Garden” Vision: Flipkart’s roadmap envisions an internal ecosystem where teams discover, deploy, and contribute AI agents, providing a practical blueprint for scalable AI agent infrastructure development.

About the Speaker

Virender Bhargav at Flipkart is a seasoned engineering leader whose expertise spans business technology integration, enterprise applications, system design/architecture, and building highly scalable systems. With a deep understanding of technology, he has spearheaded teams, modernized technology landscapes, and managed core platform layers and strategic products. With extensive experience driving innovation at companies like Paytm and Flipkart, his contributions have left a lasting impact on the industry.

Scaling Generative Models at Scale with Ray and PyTorch

Generative image models like Stable Diffusion have opened up exciting possibilities for personalization, creativity, and scalable deployment. However, fine-tuning them in production‐grade settings poses challenges: managing compute, hyperparameters, model size, data, and distributed coordination are nontrivial.

In this talk, we’ll dive deep into learning how to fine-tune Stable Diffusion models using Ray Train (with HuggingFace Diffusers), including approaches like DreamBooth and LoRA. We’ll cover what works (and what doesn’t) in scaling out training jobs, handling large data, optimizing for GPU memory and speed, and validating outputs. Attendees will come away with practical insights and patterns they can use to fine-tune generative models in their own work.

About the Speaker

Suman Debnath is a Technical Lead (ML) at Anyscale, where he focuses on distributed training, fine-tuning, and inference optimization at scale on the cloud. His work centers around building and optimizing end-to-end machine learning workflows powered by distributed computing framework like Ray, enabling scalable and efficient ML systems. Suman’s expertise spans Natural Language Processing (NLP), Large Language Models (LLMs), and Retrieval-Augmented Generation (RAG). Earlier in his career, he developed performance benchmarking and monitoring tools for distributed storage systems. Beyond engineering, Suman is an active community contributor, having spoken at over 100 global conferences and events, including PyCon, PyData, ODSC, AIE and numerous meetups worldwide.

Privacy-preserving in Computer Vision through Optics Learning

Cameras are now ubiquitous, powering computer vision systems that assist us in everyday tasks and critical settings such as operating rooms. Yet, their widespread use raises serious privacy concerns: traditional cameras are designed to capture high-resolution images, making it easy to identify sensitive attributes such as faces, nudity, or personal objects. Once acquired, such data can be misused if accessed by adversaries. Existing software-based privacy mechanisms, such as blurring or pixelation, often degrade task performance and leave vulnerabilities in the processing pipeline.

In this talk, we explore an alternative question: how can we preserve privacy before or during image acquisition? By revisiting the image formation model, we show how camera optics themselves can be learned and optimized to acquire images that are unintelligible to humans yet remain useful for downstream vision tasks like action recognition. We will discuss recent approaches to learning camera lenses that intentionally produce privacy-preserving images, blurry and unrecognizable to the human eye, but still effective for machine perception. This paradigm shift opens the door to a new generation of cameras that embed privacy directly into their hardware design.

About the Speaker

Carlos Hinojosa is a Postdoctoral researcher at King Abdullah University of Science and Technology (KAUST) working with Prof. Bernard Ghanem. His research interests span Computer Vision, Machine Learning, AI Safety, and AI for Science. He focuses on developing safe, accurate, and efficient vision systems and machine-learning models that can reliably perceive, understand, and act on information, while ensuring robustness, protecting privacy, and aligning with societal values.

It's a (Blind) Match! Towards Vision-Language Correspondence without Parallel Data

Can we match vision and language embeddings without any supervision? According to the platonic representation hypothesis, as model and dataset scales increase, distances between corresponding representations are becoming similar in both embedding spaces. Our study demonstrates that pairwise distances are often sufficient to enable unsupervised matching, allowing vision-language correspondences to be discovered without any parallel data.

About the Speaker

Dominik Schnaus is a third-year Ph.D. student in the Computer Vision Group at the Technical University of Munich (TUM), supervised by Daniel Cremers. His research centers on multimodal and self-supervised learning with a special emphasis on understanding similarities across embedding spaces of different modalities.

Oct 30 - AI, ML and Computer Vision Meetup

Join the virtual Meetup to hear talks from experts on cutting-edge topics across AI, ML, and computer vision.

Date, Time and Location

Oct 30, 2025 9 AM Pacific Online. Register for the Zoom!

The Agent Factory: Building a Platform for Enterprise-Wide AI Automation

In this talk we will explore what it takes to build an enterprise-ready AI automation platform at scale. The topics covered will include:

  • The Scale Challenge: E-commerce environments expose the limitations of single-point AI solutions, which create fragmented ecosystems lacking cohesion and efficient resource sharing across complex, knowledge-based work.
  • Root Cause Analysis Success: Flipkart’s initial AI agent transformed business analysis from days-long investigations to near-instantaneous insights, proving the concept while revealing broader platform opportunities.
  • Platform Strategy Evolution: Success across Engineering (SDLC, SRE), Operations, and Commerce teams necessitated a unified, multi-tenant platform serving diverse use cases with consistency and operational efficiency.
  • Architectural Foundation: Leveraging framework-agnostic design principles we were able to emphasize modularity, which enabled teams to leverage different AI models while maintaining consistent interfaces and scalable infrastructure.
  • The “Agent Garden” Vision: Flipkart’s roadmap envisions an internal ecosystem where teams discover, deploy, and contribute AI agents, providing a practical blueprint for scalable AI agent infrastructure development.

About the Speaker

Virender Bhargav at Flipkart is a seasoned engineering leader whose expertise spans business technology integration, enterprise applications, system design/architecture, and building highly scalable systems. With a deep understanding of technology, he has spearheaded teams, modernized technology landscapes, and managed core platform layers and strategic products. With extensive experience driving innovation at companies like Paytm and Flipkart, his contributions have left a lasting impact on the industry.

Scaling Generative Models at Scale with Ray and PyTorch

Generative image models like Stable Diffusion have opened up exciting possibilities for personalization, creativity, and scalable deployment. However, fine-tuning them in production‐grade settings poses challenges: managing compute, hyperparameters, model size, data, and distributed coordination are nontrivial.

In this talk, we’ll dive deep into learning how to fine-tune Stable Diffusion models using Ray Train (with HuggingFace Diffusers), including approaches like DreamBooth and LoRA. We’ll cover what works (and what doesn’t) in scaling out training jobs, handling large data, optimizing for GPU memory and speed, and validating outputs. Attendees will come away with practical insights and patterns they can use to fine-tune generative models in their own work.

About the Speaker

Suman Debnath is a Technical Lead (ML) at Anyscale, where he focuses on distributed training, fine-tuning, and inference optimization at scale on the cloud. His work centers around building and optimizing end-to-end machine learning workflows powered by distributed computing framework like Ray, enabling scalable and efficient ML systems. Suman’s expertise spans Natural Language Processing (NLP), Large Language Models (LLMs), and Retrieval-Augmented Generation (RAG). Earlier in his career, he developed performance benchmarking and monitoring tools for distributed storage systems. Beyond engineering, Suman is an active community contributor, having spoken at over 100 global conferences and events, including PyCon, PyData, ODSC, AIE and numerous meetups worldwide.

Privacy-preserving in Computer Vision through Optics Learning

Cameras are now ubiquitous, powering computer vision systems that assist us in everyday tasks and critical settings such as operating rooms. Yet, their widespread use raises serious privacy concerns: traditional cameras are designed to capture high-resolution images, making it easy to identify sensitive attributes such as faces, nudity, or personal objects. Once acquired, such data can be misused if accessed by adversaries. Existing software-based privacy mechanisms, such as blurring or pixelation, often degrade task performance and leave vulnerabilities in the processing pipeline.

In this talk, we explore an alternative question: how can we preserve privacy before or during image acquisition? By revisiting the image formation model, we show how camera optics themselves can be learned and optimized to acquire images that are unintelligible to humans yet remain useful for downstream vision tasks like action recognition. We will discuss recent approaches to learning camera lenses that intentionally produce privacy-preserving images, blurry and unrecognizable to the human eye, but still effective for machine perception. This paradigm shift opens the door to a new generation of cameras that embed privacy directly into their hardware design.

About the Speaker

Carlos Hinojosa is a Postdoctoral researcher at King Abdullah University of Science and Technology (KAUST) working with Prof. Bernard Ghanem. His research interests span Computer Vision, Machine Learning, AI Safety, and AI for Science. He focuses on developing safe, accurate, and efficient vision systems and machine-learning models that can reliably perceive, understand, and act on information, while ensuring robustness, protecting privacy, and aligning with societal values.

It's a (Blind) Match! Towards Vision-Language Correspondence without Parallel Data

Can we match vision and language embeddings without any supervision? According to the platonic representation hypothesis, as model and dataset scales increase, distances between corresponding representations are becoming similar in both embedding spaces. Our study demonstrates that pairwise distances are often sufficient to enable unsupervised matching, allowing vision-language correspondences to be discovered without any parallel data.

About the Speaker

Dominik Schnaus is a third-year Ph.D. student in the Computer Vision Group at the Technical University of Munich (TUM), supervised by Daniel Cremers. His research centers on multimodal and self-supervised learning with a special emphasis on understanding similarities across embedding spaces of different modalities.

Oct 30 - AI, ML and Computer Vision Meetup

Join the virtual Meetup to hear talks from experts on cutting-edge topics across AI, ML, and computer vision.

Date, Time and Location

Oct 30, 2025 9 AM Pacific Online. Register for the Zoom!

The Agent Factory: Building a Platform for Enterprise-Wide AI Automation

In this talk we will explore what it takes to build an enterprise-ready AI automation platform at scale. The topics covered will include:

  • The Scale Challenge: E-commerce environments expose the limitations of single-point AI solutions, which create fragmented ecosystems lacking cohesion and efficient resource sharing across complex, knowledge-based work.
  • Root Cause Analysis Success: Flipkart’s initial AI agent transformed business analysis from days-long investigations to near-instantaneous insights, proving the concept while revealing broader platform opportunities.
  • Platform Strategy Evolution: Success across Engineering (SDLC, SRE), Operations, and Commerce teams necessitated a unified, multi-tenant platform serving diverse use cases with consistency and operational efficiency.
  • Architectural Foundation: Leveraging framework-agnostic design principles we were able to emphasize modularity, which enabled teams to leverage different AI models while maintaining consistent interfaces and scalable infrastructure.
  • The “Agent Garden” Vision: Flipkart’s roadmap envisions an internal ecosystem where teams discover, deploy, and contribute AI agents, providing a practical blueprint for scalable AI agent infrastructure development.

About the Speaker

Virender Bhargav at Flipkart is a seasoned engineering leader whose expertise spans business technology integration, enterprise applications, system design/architecture, and building highly scalable systems. With a deep understanding of technology, he has spearheaded teams, modernized technology landscapes, and managed core platform layers and strategic products. With extensive experience driving innovation at companies like Paytm and Flipkart, his contributions have left a lasting impact on the industry.

Scaling Generative Models at Scale with Ray and PyTorch

Generative image models like Stable Diffusion have opened up exciting possibilities for personalization, creativity, and scalable deployment. However, fine-tuning them in production‐grade settings poses challenges: managing compute, hyperparameters, model size, data, and distributed coordination are nontrivial.

In this talk, we’ll dive deep into learning how to fine-tune Stable Diffusion models using Ray Train (with HuggingFace Diffusers), including approaches like DreamBooth and LoRA. We’ll cover what works (and what doesn’t) in scaling out training jobs, handling large data, optimizing for GPU memory and speed, and validating outputs. Attendees will come away with practical insights and patterns they can use to fine-tune generative models in their own work.

About the Speaker

Suman Debnath is a Technical Lead (ML) at Anyscale, where he focuses on distributed training, fine-tuning, and inference optimization at scale on the cloud. His work centers around building and optimizing end-to-end machine learning workflows powered by distributed computing framework like Ray, enabling scalable and efficient ML systems. Suman’s expertise spans Natural Language Processing (NLP), Large Language Models (LLMs), and Retrieval-Augmented Generation (RAG). Earlier in his career, he developed performance benchmarking and monitoring tools for distributed storage systems. Beyond engineering, Suman is an active community contributor, having spoken at over 100 global conferences and events, including PyCon, PyData, ODSC, AIE and numerous meetups worldwide.

Privacy-preserving in Computer Vision through Optics Learning

Cameras are now ubiquitous, powering computer vision systems that assist us in everyday tasks and critical settings such as operating rooms. Yet, their widespread use raises serious privacy concerns: traditional cameras are designed to capture high-resolution images, making it easy to identify sensitive attributes such as faces, nudity, or personal objects. Once acquired, such data can be misused if accessed by adversaries. Existing software-based privacy mechanisms, such as blurring or pixelation, often degrade task performance and leave vulnerabilities in the processing pipeline.

In this talk, we explore an alternative question: how can we preserve privacy before or during image acquisition? By revisiting the image formation model, we show how camera optics themselves can be learned and optimized to acquire images that are unintelligible to humans yet remain useful for downstream vision tasks like action recognition. We will discuss recent approaches to learning camera lenses that intentionally produce privacy-preserving images, blurry and unrecognizable to the human eye, but still effective for machine perception. This paradigm shift opens the door to a new generation of cameras that embed privacy directly into their hardware design.

About the Speaker

Carlos Hinojosa is a Postdoctoral researcher at King Abdullah University of Science and Technology (KAUST) working with Prof. Bernard Ghanem. His research interests span Computer Vision, Machine Learning, AI Safety, and AI for Science. He focuses on developing safe, accurate, and efficient vision systems and machine-learning models that can reliably perceive, understand, and act on information, while ensuring robustness, protecting privacy, and aligning with societal values.

It's a (Blind) Match! Towards Vision-Language Correspondence without Parallel Data

Can we match vision and language embeddings without any supervision? According to the platonic representation hypothesis, as model and dataset scales increase, distances between corresponding representations are becoming similar in both embedding spaces. Our study demonstrates that pairwise distances are often sufficient to enable unsupervised matching, allowing vision-language correspondences to be discovered without any parallel data.

About the Speaker

Dominik Schnaus is a third-year Ph.D. student in the Computer Vision Group at the Technical University of Munich (TUM), supervised by Daniel Cremers. His research centers on multimodal and self-supervised learning with a special emphasis on understanding similarities across embedding spaces of different modalities.

Oct 30 - AI, ML and Computer Vision Meetup
Event DataTalks.Club 2025-10-10
Ranjitha Kulkarni – Machine Learning Engineer @ NeuBird AI (past: Microsoft, Dropbox)

In this episode, we talked with Ranjitha Kulkarni, a machine learning engineer with a rich career spanning Microsoft, Dropbox, and now NeuBird AI. Ranjitha shares her journey into ML and NLP, her work building recommendation systems, early AI agents, and cutting-edge LLM-powered products. She offers insights into designing reliable AI systems in the new era of generative AI and agents, and how context engineering and dynamic planning shape the future of AI products.TIMECODES00:00 Career journey and early curiosity04:25 Speech recognition at Microsoft05:52 Recommendation systems and early agents at Dropbox07:44 Joining NewBird AI12:01 Defining agents and LLM orchestration16:11 Agent planning strategies18:23 Agent implementation approaches22:50 Context engineering essentials30:27 RAG evolution in agent systems37:39 RAG vs agent use cases40:30 Dynamic planning in AI assistants43:00 AI productivity tools at Dropbox46:00 Evaluating AI agents53:20 Reliable tool usage challenges58:17 Future of agents in engineering Connect with Ranjitha- Linkedin - https://www.linkedin.com/in/ranjitha-gurunath-kulkarniConnect with DataTalks.Club:- Join the community - https://datatalks.club/slack.html- Subscribe to our Google calendar to have all our events in your calendar - https://calendar.google.com/calendar/r?cid=ZjhxaWRqbnEwamhzY3A4ODA5azFlZ2hzNjBAZ3JvdXAuY2FsZW5kYXIuZ29vZ2xlLmNvbQ- Check other upcoming events - https://lu.ma/dtc-events- GitHub: https://github.com/DataTalksClub- LinkedIn - https://www.linkedin.com/company/datatalks-club/ - Twitter - https://twitter.com/DataTalksClub - Website - https://datatalks.club/

AI/ML GenAI GitHub HTML LLM Microsoft NLP RAG
Abouzar Abbaspour – data engineer @ Tesla

In this episode, we talked with Abouzar Abbaspour, a data engineer whose career spans software engineering in Iran, building crowd and recommendation systems at a Dutch theme park, deploying large-scale ML models at Bol.com, and now working at Tesla. Abouzar shares how he bridged diverse industries, tackled real-world data challenges, and adapted to new roles while keeping a hands-on approach to machine learning and engineering.TIMECODES00:00 Career journey and early motivations06:17 Moving to Europe for data science12:18 Working with theme parks and crowd modeling18:29 Lessons from ride and visitor data23:06 Building recommendation systems at Efteling27:26 Joining Bol.com and the Dutch e-commerce industry32:49 Product and brand recommendation logic36:09 Experimenting with "Tinder for brands"40:26 Engagement metrics and product validation43:02 From ML engineering to data engineering roles52:04 Hands-on skills at Tesla and industry expectations57:43 Career growth, learning, and adviceConnect with AbouzarLinkedin -   / abouzar-abbaspour   Website - https://www.abouzar-abbaspour.com/ Connect with DataTalks.Club: Join the community - https://datatalks.club/slack.htmlSubscribe to our Google calendar to have all our events in your calendar - https://calendar.google.com/calendar/...Check other upcoming events - https://lu.ma/dtc-eventsGitHub: https://github.com/DataTalksClubLinkedIn -   / datatalks-club   Twitter -   / datatalksclub   Website - https://datatalks.club/

AI/ML Data Engineering GitHub
Irina Malkova – VP Data & AI @ Salesforce , Joe Reis – founder @ Ternary Data

Are dashboards dead? For complex enterprise use cases, the answer might be yes. In this episode, I'm joined by Irina Malkova (VP Data & AI at Salesforce), to discuss her team's transformational journey from building complex dashboards to deploying AI-powered conversational agents. We dive deep into how this shift is not just a change in tooling, but a fundamental change in how users access insights and how data teams measure their impact.

Join us as we cover: The Shift from Dashboards to Agents: We discuss why dashboards can create a high cognitive load and fail users in complex scenarios , and how conversational agents in the flow of work (like Slack) provide targeted, actionable insights and boost adoption.What is Product Telemetry?: Irina explains how telemetry is evolving from a simple engineering observability use case to a critical data source for AI, machine learning, and recommendation systems.Why Standard RAG Fails in the Enterprise: Irina shares why typical RAG approaches break down on dense, entity-rich corporate data (like Salesforce's help docs) where semantic similarity isn't enough, leading to the rise of Graph RAG.The New, Measurable ROI of Data: How moving from BI to agents allows data teams to precisely measure impact, track downstream actions, and finally have a concrete answer to the ROI question that was previously impossible to justify.Data Teams as Enterprise Leaders: Why data teams are uniquely positioned to lead AI transformation, as they hold the enterprise "ontology" and have experience building products under uncertainty.

AI/ML BI RAG
The Joe Reis Show

Ranjitha has shipped AI features across multiple generations of NLP: from speech recognition and online evaluation at Microsoft to LLMs, RAG, and agents in production at Dropbox Dash, and now agentic systems at NeuBird.ai. In this conversation, she shares the concrete practices that make assistants useful, trustworthy, and maintainable in real products. ​ ​We plan to cover:

  • ​How early work in speech systems still shapes today’s LLMs and agents
  • ​What it takes to turn an LLM demo into a dependable product
  • ​Where RAG and agents shine and where they fall short
  • ​Skills engineers need today to succeed with applied NLP and agents

​ ​About the speaker

Ranjitha Gurunath Kulkarni is a Staff Machine Learning Engineer at NeuBird.ai. Previously, she built LLM- and agent-powered product capabilities at Dropbox Dash and worked on speech recognition, language modeling, online metrics, and assistant evaluation at Microsoft. Her publications span voice query reformulation and automatic online evaluation of intelligent assistants, and her patents include automated closed captioning using temporal data and hyperarticulation detection. Ranjitha holds a master’s from Carnegie Mellon University (Language Technologies Institute). ​

Join our slack: https://datatalks.club/slack.html

Building reliable AI products in the era of Gen AI and Agents

A conversation with Abouzar Abbaspour on ML engineering, data pipelines, and real-world impact ​ ​What does it take to turn raw data into products that millions of people love to use? Abouzar Abbaspour will discuss this based on his experience working with startups, amusement parks, e-commerce, and now Tesla.

​At Efteling theme park, he built queue-time forecasting and recommendation systems for visitors. At bol.com, he helped deploy a recommendation engine to over 6 million users. And today at Tesla, he works on predictive maintenance and integrates LLM agents into production systems.

​In this conversation, Abouzar shares stories from the front lines of data and ML engineering: what succeeds, what fails, and what matters when you move from experiments to production. ​ ​We plan to cover:

  • ​Forecasting queues and building recommendation systems at a theme park
  • ​Deploying large-scale recommender systems at bol.com
  • ​The leap from data engineering into ML engineering
  • ​Predictive maintenance and LLM agents at Tesla
  • ​Why productionizing ML is about much more than the model
  • ​Which trends in data and ML are hype, and which are here to stay

​ ​About the speaker

Abouzar Abbaspour is a machine learning and data engineer whose career spans startups, academia, e-commerce, theme parks, and automotive. He co-founded a telecom startup in Iran, worked on forecasting models and recommendation engines at Efteling, and later deployed large-scale ML models at bol.com. Today, at Tesla, he focuses on predictive maintenance, LLM agents, and scalable pipelines. Abouzar holds an EngD in Data Science from Eindhoven University of Technology.

Join our slack: https://datatalks.club/slack.html

From Theme Parks to Tesla: Building Data Products That Work

​Micheal Lanham has been building intelligent systems since the early 2000s, starting with neural networks and evolutionary algorithms in games and moving through enterprise software, geoscience, AR/VR, and now AI agents. Over the years, he has written more than ten technical books and worked across industries as an architect, manager, and hands-on AI engineer.

​In this conversation, Micheal shares hard-earned lessons from two decades at the intersection of data, software, and AI. We’ll explore what games can teach us about intelligence, why evolutionary methods are resurfacing, and how to think about AI agents beyond the hype. ​ ​We plan to cover:

  • ​What games can teach us about AI and data products
  • ​The promise (and limits) of evolutionary deep learning
  • ​AI agents in practice: beyond LLMs and prompts
  • ​How XR and intelligent systems are converging
  • ​What it takes to productionize AI across industries
  • ​What has really changed in AI over the last 20 years

​ ​About the speaker

Micheal Lanham is a best-selling author, innovator, and AI engineer based in Calgary, Canada. His work spans games, graphics, GIS, enterprise software, and machine learning. He has published over ten technical books, including Evolutionary Deep Learning, Hands-On Reinforcement Learning for Games, and AI Agents in Action. Micheal has worked as a lead AI developer, architect, and manager across industries from oil and gas to fintech, and today focuses on building intelligent systems with deep reinforcement learning, evolutionary methods, and generative AI.

Join our slack: https://datatalks.club/slack.html

Lessons from Two Decades of AI