Retrieval-augmented generation (RAG) is one of the best (and easiest) ways to specialize an LLM over your own data, but successfully applying RAG in practice involves more than just stitching together pretrained models.
Ben and Eira talk with LlamaIndex CEO and cofounder Jerry Liu, along with venture capitalist Jerry Chen, about how the company is making it easier for developers to build LLM apps. They touch on the importance of high-quality training data to improve accuracy and relevance, the role of prompt engineering, the impact of larger context windows, and the challenges of setting up retrieval-augmented generation (RAG).
In this episode, Ben chats with Elastic software engineering director Paul Oremland along with Stack Overflow staff software engineer Steffi Grewenig and senior software developer Gregor Časar about vector databases and semantic search from both the vendor and customer perspectives.
Here’s a simple, three-part framework that explains generative language models.
It’s easy to generate code, but not so easy to generate good code.
Ben and Ryan are joined by Robin Gupta for a conversation about benchmarking and testing AI systems. They talk through the lack of trust and confidence in AI, the inherent challenges of nondeterministic systems, the role of human verification, and whether we can (or should) expect an AI to be reliable.
Ben and Ryan talk with Vikram Chatterji, founder and CEO of Galileo, a company focused on building and evaluating generative AI apps. They discuss the challenges of benchmarking and evaluating GenAI models, the importance of data quality in AI systems, and the trade-offs between using pre-trained models and fine-tuning models with custom data.
Product manager Ash Zade joins the home team to talk about the journey to OverflowAI, a GenAI-powered add-on for Stack Overflow for Teams that’s available now. Ash describes how his team built Enhanced Search, the problems they set out to solve, how they ensured data quality and accuracy, the role of metadata and prompt engineering, and the feedback they’ve gotten from users so far.
Only about 5% of GenAI projects lead to significant monetization of new product offerings.
The home team talks about the current state of the software job market, the changing sentiments around AI job opportunities, the impact of big players like Facebook and OpenAI on the space, and the challenges for startups. Plus: The philosophical implications of LLMs and the friendship potential of corvids.
Ben talks with Shane McAllister, lead developer advocate at MongoDB, Stanimira Vlaeva, senior developer advocate at MongoDB, and Miku Jha, director, AI/ML and generative AI at Google Cloud, about the challenges and opportunities of operationalizing and scaling generative AI models in enterprise organizations.
On this episode: Stack Overflow senior data scientist Michael Geden tells Ryan and Ben about how data scientists evaluate large language models (LLMs) and their output. They cover the challenges involved in evaluating LLMs, how LLMs are being used to evaluate other LLMs, the importance of data validating, the need for human raters, and more needs and tradeoffs involved in selecting and fine-tuning LLMs.
In this sponsored episode, Ben and Ryan are joined by Ria Cheruvu, an AI evangelist at Intel, to discuss the different approaches to incorporating AI models into organizations.
The home team convenes to discuss the XZ backdoor attack, what great software engineers have in common, how GenAI is changing the face of drug development, and the rise of managed service providers for AI.
We sit down with Jessica Clark, a senior data scientist at Stack Overflow, to discuss how our company approaches generative AI and data quality.
This new LLM technique has started improving the results of models without additional training.
Ben and Ryan are joined by Nick Heudecker, Senior Director of Market Strategy and Competitive Intelligence at Cribl, to discuss the state of data and analytics. They cover GenAI, the role of incumbents vs. startups, challenges of data storage and security, data quality and ETL pipelines, measures of data quality for GenAI, and Cribl’s role in the data and observability space.
Ben and Ryan are joined by Bill Harding, CEO of GitClear, for a discussion of AI-generated code quality and its impact on productivity. GitClear’s research has highlighted the fact that while AI can suggest valid code, it can’t necessarily reuse and modify existing code—a recipe for long-term challenges in maintainability and test coverage if devs are too dependent on AI code-gen tools.
The home team discusses the challenges (hardware and otherwise) of building AI models at scale, why major players like Meta are open-sourcing their AI projects, what Apple’s recent changes mean for developers in the EU, and Perplexity AI’s new approach to search.
Machine learning scientist, author, and LLM developer Maxime Labonne talks with Ben and Ryan about his role as lead machine learning scientist, his contributions to the open-source community, the value of retrieval-augmented generation (RAG), and the process of fine-tuning and unfreezing layers in LLMs. The team talks through various challenges and considerations in implementing GenAI, from data quality to integration.
Stack Overflow is on a journey to build a new era in the practice of AI: the era of social responsibility. All products based on models that consume public Stack Overflow data are required to provide attribution back to the highest-relevance posts that influenced the summary given by the model.
Ryan and Ben chat with Raymond Lo, AI software evangelist at Intel, about the AI PC, the software that powers AI breakthroughs, and optimizing hardware and software in unison to improve generative AI performance.
If you’re building experimental GenAI features that haven’t proven their product market fit, you don’t want to commit to a model that runs up costs without a return on that investment.
On this home team episode: Massachusetts makes a welcome shift toward skills-based hiring, AI-generated content robs us of our appetite for mac and cheese, and large-scale crypto mining operations account for more than 2% of the US’s electricity generation. Plus: A PDF quite a bit bigger than Germany.
