Matt Welsh on AI-Assisted Programming, Natural Language Interfaces, Advanced Data Extraction, and Trust in AI Systems.
Subscribe: Apple • Spotify • Overcast • Pocket Casts • AntennaPod • Podcast Addict • Amazon • RSS.
Matt Welsh is a technical leader at Aryn AI, an AI-powered ETL system for RAG frameworks, LLM-based applications, and vector databases. In this episode, we explore how AI is revolutionizing programming and software development. We discuss AI-assisted tools like GitHub Copilot and ChatGPT, the rise of natural language programming, and AI’s expanding role in code review and operating systems. We also address challenges in trust and verification of AI-generated code, advanced data extraction techniques, and innovative frameworks for unstructured analytics. Finally, we highlight the importance of cross-functional collaboration in building trustworthy AI systems and the future of user interfaces, including voice-based interactions. [This episode originally aired on Generative AI in the Real World, a podcast series I’m hosting for O’Reilly.]
Interview highlights – key sections from the video version:
- The Changing Nature of Programming
- Trusting AI to Build Pipelines
- The Democratization of Programming
- The Role of LLMs in Programming
- Challenges in Integrating LLMs into Workflows
- The Future of Computing Interfaces & Real-Time Speech Synthesis
- How Programmers Are Using AI
- The Challenges of Pandas and ETL for RAG Frameworks
- The Importance of Code Reviews and Testing in an AI-Driven World
- The Potential of AI Mentorship for Junior Developers & Data Quality
- Automating Code Review & Aryn: A Framework for Unstructured Data
- The Importance of Information Extraction and Ranking in RAG
- The Limitations of RAG, Structured Queries, & Knowledge Graphs
- The Challenge of Evaluating AI Systems & The Need for Collaboration
Related content:
- A video version of this conversation is available on our YouTube channel.
- Is Your Data Strategy Ready for Generative AI?
- Generative AI: Navigating the Challenges of Enterprise Adoption
- Unpacking Model Collaboration: Ensembles, Routers, and Merging
- Shuveb Hussain → Automating Unstructured Data Extraction with LLMs
- Philip Rathle → Supercharging AI with Graphs
- Brian Raymond → ETL for LLMs
- Jerry Liu → An Open Source Data Framework for LLMs
- Andrew Ng → Advancing AI: Scaling, Data, Agents, Testing, and Ethical Considerations
If you enjoyed this episode, please support our work by encouraging your friends and colleagues to subscribe to our newsletter:
