talk-data.com
People (2 results)
Activities & events
| Title & Speakers | Event |
|---|---|
|
Proof of Concept or Proof of Chaos? Develop AI POC that impress and create value for your product
2025-06-17 · 19:40
Raffaele Bongo
– Senior ML Engineer & Data Solutions Architect
@ Talentware
With hundreds of new AI technologies and methodologies emerging daily, having a robust and holistic Proof of Concept (POC) framework to identify what truly adds value to a product has never been more critical. POC projects are often underestimated, yet they play a pivotal role in the success of AI initiatives and features. This talk highlights the importance of bringing rigor back to AI POCs, offering a structured, end-to-end approach to ensure their effectiveness. It also examines how AI feature POCs intersect with traditional software and design processes, clarifying AI’s role within the broader product development lifecycle. We’ll explore common pitfalls and strategies to address them. You’ll learn how to: 1. Define scope: Set clear objectives, boundaries, and success metrics while involving the right stakeholders from the start. 2. Iterate effectively: Track experiments properly, build in a flexible and repeatable way, and be ready to pivot or start over when necessary. 3. Facilitate feedback loops: Present results effectively, collect actionable feedback, and guide discussions to improve the solution. By applying these principles, teams can develop AI POCs that not only prove value but also lay the groundwork for scalable, production-ready solutions. |
|
|
Stock Market Forecasting with Nixtla Conformal Prediction and Stock Agent Analysis
2025-06-17 · 19:00
Claudio Giorgio Giancaterino
– Actuary & AI Scientist
@ Intesa San Paolo Assicurazioni
Dive into the future of financial analysis and the power of uncertainty quantification using conformal prediction techniques, applied through the robust capabilities of Nixtla library. Gain insights into how this tool provides reliable predictive intervals, enhancing decision-making in volatile markets. Advantages and Fundamentals concepts of Conformal Prediction: Explore the impact of conformal prediction on financial time series forecasting. This talk will introduce the fundamentals of conformal prediction, highlighting its ability to provide reliable uncertainty quantification by generating predictive intervals that encompass real market behaviours. In this way conformal prediction allows investors to make informed decisions despite inherent market volatility. Nixtla library and practical example of a financial time series: To bring theory into practice, Claudio will present the Nixtla library, which simplifies the implementation of conformal prediction, and walk us through a use case using Yahoo! Finance's API with a comparison of TimeGPT against several statistical and machine learning models. Streamlit App: Develop a Streamlit App for a general Stock Market Forecasting, supported by a Stock Agent Analysis built with pyautogen |
|
|
Synthetic Data Assessment and Best Practices for Data Department Efficacy
2025-01-22 · 17:30
📅 Date: 22-01-2025 📍 Location: Talentware @ WAO PL7, Via Luigi Porro Lambertenghi 7 (Isola/Zara/Garibaldi FS) Agenda
Join us for dinner after the event too to continue chatting! Talks and Speakers SURE: A New Privacy and Utility Assessment Library for Synthetic Data In this talk, we will explore the core principles of validating and evaluating synthetic data across diverse domains, highlighting the importance of scalable and robust solutions. We will introduce SURE, an open-source library designed to assess privacy risks and utility trade-offs in synthetic datasets. By showcasing how SURE leverages the power of Polars to efficiently generate detailed and insightful reports, we’ll demonstrate its role in streamlining synthetic data assessment. Finally, we will discuss the current open challenges in its development and future directions. Dario Brunelli ventured into the world of data with a Master’s in Quantum Machine Learning, leveraging his M.Sc. in Electrical Engineering and a dynamic background in motorsport. As a Machine Learning Engineer at Clearbox AI, he specializes in synthetic data generation, with a focus on data privacy and time-series generation. When he's not coding, Dario enjoys photography, cycling, and surfing. Luca Gilli is the co-founder of Clearbox AI, a company specializing in synthetic data generation solutions to drive innovation and enhance data privacy. He holds a PhD in computational mathematics from Delft University of Technology. He worked as a scientific software developer for a consultancy firm for over 5 years in the Netherlands. Outside of work, he enjoys hiking in the beautiful Piedmontese mountains. He’s also passionate about farming, with a special interest in experimenting with hydroponic techniques. Going beyond traditional software playbook to build an effective data department The journey to AI maturity is full of challenges. In Italy, while 65-70% of companies have embraced digitalization and 60% utilize cloud computing, only 9-11% have achieved mature AI solutions. This disparity underscores the evolving nature of AI as a discipline and its distinction from traditional software development. With over 80% of AI projects failing due to unclear expectations, inadequate frameworks, and communication gaps, the need for a structured approach has never been more urgent. This talk introduces the foundational pillars of Talentware’s Data Department that is:
By addressing these pillars, the session highlights how AI projects differ from traditional software, and that their success heavily depends on all professionals involved at every level understanding this distinction. Raffaele Bongo is a Senior Machine Learning Engineer and Data Solutions Architect at Talentware. Raffaele has an extensive experience in startups and scale-ups, where he has specialized in building and deploying machine learning models and data pipelines on cloud platforms. Originally, Raffaele signed up for mechanical engineering but switched to computer science last minute because AI sounded cooler. Great call! |
Synthetic Data Assessment and Best Practices for Data Department Efficacy
|