MLOps.community

Egor Kraev, Principal AI Scientist at Wise Plc and founder of the Swiss Pirate Party, dives into the transformative power of AI in fintech. He shares insights on integrating large language models into machine learning pipelines and the practical implications of his open-source MotleyCrew framework. Egor highlights the role of AI in improving fraud detection and optimizing currency flow. He also discusses the importance of autonomy within teams, navigating causal inference in marketing, and enhancing user engagement through targeted campaigns.

In this discussion, Michelle Marie Conway, Lead Data Scientist at Lloyds Banking Group, and Andrew Baker, Data Science Delivery Lead, share insights from their cloud migration journey. They delve into the transition from on-prem technology to the cloud, highlighting the complexities of model management and engineering practices. Their conversation also touches on the harmony between music and technology, the challenges of chaos engineering in regulated environments, and the importance of collaboration within data science and platform teams.

In this insightful discussion, Jineet Doshi, an award-winning AI lead with over seven years at Intuit, dives deep into the complexities of evaluating generative AI systems. He emphasizes the importance of holistic evaluation to foster trust and the unique challenges posed by large language models. Jineet explores diverse evaluation methods, from classic NLP techniques to innovative strategies like red teaming. He also tackles the financial nuances of generative AI and the balance between human insight and automated feedback for robust assessments.

Robert Caulk, the founder of Emergent Methods and an expert in large-scale applications, discusses the cutting-edge development of unconstrained knowledge graphs to counter misinformation. He reveals how new tools allow for the processing of vast amounts of news data more efficiently. The podcast explores the integration of knowledge graphs with AI, enhancing user interaction and the fight against false narratives. Caulk emphasizes the ethical challenges of data handling and the role of advanced AI models in improving sentiment analysis, showcasing a future of responsible information management.

Guanhua Wang, a Senior Researcher in the DeepSpeed team at Microsoft, dives into the revolutionary Domino training engine, designed to eliminate communication overhead during LLM training. He discusses the intricacies of naming the Phi-3 model and the growing interest in smaller language models. Wang highlights advanced techniques like data offloading and quantization, showcasing how Domino can speed up training by up to 1.3x compared to existing methods, while addressing privacy in customizable copilot models. It's a deep dive into optimizing AI training!

Aditya Naganath, an experienced investor at Kleiner Perkins, delves into AI's next frontier, focusing on the collaboration between AI and knowledge workers. He discusses the evolving landscape of AI investments, emphasizing the significance of strong teams and go-to-market strategies. The conversation also highlights the shift towards low-code and no-code tools, democratizing access to technology, and innovative challenges in AI infrastructure. Aditya provides insights into GPU reliability issues, economic dynamics in AI services, and the growing importance of inference in the AI space.

Vincent Moens, an Applied Machine Learning Research Scientist at Meta and the author behind TorchRL and TensorDict, delves into the fascinating applications of PyTorch in control systems and decision-making. He shares insights on optimizing performance using practical tips, including the nuances of pin memory for CUDA transfers. The discussion covers the pitfalls of in-place tensor modifications and introduces TensorDict as a solution for efficient data handling. Additionally, Vincent emphasizes community collaboration to enhance developer experiences and improve user-friendly APIs in PyTorch.

In this engaging discussion, Matt van Itallie, founder and CEO of Sema, shares insights on the importance of comprehensive codebase scans for technical due diligence. He reveals how Generative AI is reshaping code transparency and introduces the Generative AI Bill of Materials (GBOM) for managing AI-generated code risks. Matt emphasizes the necessity of bridging technical and business viewpoints in AI evaluation, highlighting practical strategies for assessing cloud costs and optimizing code quality. His insights are invaluable for both technical and non-technical audiences.

Michael Gschwind, Director/Principal Engineer for PyTorch at Meta Platforms, shares his insights on AI advancements. He discusses the evolution from gaming hardware to modern AI, highlighting the pivotal role of community collaboration. The conversation covers the development of Torch Chat for large language models, energy-efficient optimization techniques, and the exciting shift toward on-device AI solutions. Gschwind also emphasizes strategic optimization to avoid premature pitfalls in technology development.

Join Meera Clark, a Principal at Redpoint Ventures, Sandeep Bakshi, from Prosus Ventures, and George Robson of Sequoia Capital as they dissect the thrilling yet challenging realm of AI investments. They explore the exciting applications of large language models in various industries, the strategic pivots needed for new startups to compete with giants, and the evolving expectations of consumers. The panel sheds light on economic hurdles, scalability issues, and the vital role of sustainable business models in navigating the future of generative AI.