Go Beyond the Full Stack. Master the Reliable Enterprise AI Stack at AI By the Bay conference.
- Oli Dinov
- Aug 1
- 3 min read
Updated: Sep 12
In 2023, we hosted the fundamentals of LLM by Full Stack Deep Learning. Now, we help developers master the Enterprise AI. Join 1,000+ senior engineers Nov 17–19 in Oakland, CA to solve the production challenges that come next.
Dates & Location: November 17–19, 2025 • Scottish Rite Center, Oakland, CA.
Audience: For Senior Engineers, Infra Leads & Platform Teams.
Why Attend?
For years, FSDL has set the standard for learning how to build real AI products, not just theoretical models. Their philosophy is simple: cover the full stack, from idea to deployment, with a focus on what works in the real world. In 2023, we were proud to host the FSDL bootcamp as part of the conference. For 2025, we're infusing that same practical, engineering-first spirit into the entire event. AI by the Bay is now the definitive gathering for practitioners who want to put the full-stack philosophy into practice at enterprise scale.
Your Next Mile
Module | Key Concepts | Next Mile at AI by the Bay | Relevant Talks/Speakers |
Learn to Spell: Prompt Engineering | High-level intuitions; “learn to spell” with prompts like chain-of-thought. | From prompt iteration to structured eval harnesses and grammar constraints for production trust. | Vaibhav Gupta on BAML, a new programming language for building maintainable, testable AI pipelines that handle complex documents with near-perfect accuracy. Can We Trust AI-Generated Code? by Baruch Sadogursky/Leonid Igolnik. |
LLMOps | Model comparison, prompt management; testing challenges from non-determinism. | From tracking prompts in Git to implementing a full CI/CD system for models, ensuring security, and choosing the right model strategy (OSS vs. proprietary). | Mihai Maruseac on Practical Model Signing with Sigstore to solve ML supply chain security. Donny Greenberg on when to fine-tune SLMs on first-party data to Beat GPT-4 for specific tasks. Matt White on navigating the open-source landscape Beyond Open Weights to make informed licensing and governance decisions. |
UX for Language User Interfaces | User-centered patterns; case studies like Copilot. | From prototypes to adaptive multi-agent interfaces with real-time storytelling. | |
Augmented Language Models | Augmenting with knowledge, vectors, tools. | From building a basic RAG prototype to engineering enterprise-grade retrieval systems with structured data, real-time streams, and scalable memory. | Danica Fine on enriching RAG with Apache Iceberg for high-performance structured retrieval. Viktor Gamov on using MCP to Bridge LLMs and Data Streams like Kafka. Vasilije Markovic on Building Memory for AI Agents at scale using graph and vector databases. |
Building with Agents | Rapid prototyping with simple stacks. | From understanding the "agent" design pattern to architecting, deploying, and trusting multi-agent systems that handle complex workflows in production. | Steve Anderson on Trusted Agentic AI for governance and explainability. Mary Grygleski on using Event-Driven and Multi-Agent Architectures for complex business workflows. Michael Maximilien on Building AI Agent Workflows Fast with no-code tools. James Ward's hands-on workshop on Building Agents with Spring AI and MCP. |
The Full Stack: Data & Infrastructure | Basics of Transformers and notable LLMs. | From understanding the Transformer architecture to navigating the complex data infrastructure required to support it at scale in a modern data lakehouse. | Julien Le Dem (co-creator of Parquet/Arrow) on The Advent of the Open Data Lake. Alex Merced on using Arrow Flight and Apache Iceberg for high-performance, interoperable AI development. Robert Nishihara (co-creator of Ray) on Scaling Multimodal Data with a modern software stack (Kubernetes, Ray, PyTorch). |