Production-ready AI Agents Hackathon @ AI By the Bay
Join us for a 2-day hackathon dedicated to building production-ready AI Agents that can accomplish something amazing in a business-to-business or business-to-consumer use case.
THE CHALLENGE
You will need to make sure your agent is maintainable, scalable, and observable and also secure by design, i.e., safe from agent-related attack vectors (prompt injection, data exfiltration, privilege escalation, broken access controls, etc).
You will need to illustrate how your agent accomplishes these production-ready tasks while still providing access to business data and processes. Bonus points if you use tools to validate the production readiness of your agent.
WHO SHOULD ATTEND
Developers looking to build their next production-ready AI agent project, alongside the best software engineers and legends within the Bay Area and beyond.
Max team size is 5. No team? No problem. We’ll help match you with others at the event. At least 30% of the team should come in person.
All attendees must be registered and approved on Luma.
In-person attendees will get complimentary expo passes to the AI By the Bay conference.
We’d be happy to help you make the case to your manager about the value of participating in our hackathon. Just email us at [email protected], and we’ll send you key points you can share.
AGENDA
Subject to change.
Tue, Nov 18
9:00 am – 10:00 am | Onboarding, registration, coffee
10:00 am – 10:30 am | Kickoff, rules, safety/CoC
12:30 pm - 1:30 pm | Lunch
Mentor office hours
2:30 pm | Checkpoint #1 (problem framing & feasibility)
4:30 pm | Checkpoint #2 (reliability/readiness review)
+ more details soon
3:30 pm - 4:00 pm | Coffee break
6:00 pm | Happy Hour
7:00 pm | Doors close
Wed, Nov 19
8:00 am | Doors open, coffee
Mentor office hours (+ more details soon)
11:00 am | Submissions due (repo + 3-min demo video)
11:00 am - 1:30 pm | Online Judging
12:30 pm - 1:30 pm | Lunch
1:30 pm - 3:00 pm | In-Person Judging (a team representative has to be at the venue)
3:00 pm – 4:10 pm | Finalist lightning demos - will be recorded and posted on our YouTube channel
4:10 pm - 4:40 pm | Coffee Break
4:40 pm – 5:10 pm | Closing panel with judges
5:10 pm – 5:20 pm | Awards + group photo
5:20 pm – 5:40 pm | Closing remarks
CREDITS
- AWS accounts will be provided enabling you to pick from numerous hosted Amazon Bedrock models and compute environments.
MENTORS & JUDGES
While participants are free to use any frameworks and tools they prefer, we will have an experienced group of mentors and judges on site to provide guidance.
James Ward known for his deep expertise in not only functional programming but also in working with scalable AI solutions in the AWS cloud. [About]
Josh Long has long been known as “Mr. Spring Boot.” More recently, he's been experimenting with Spring AI ever since Rod Johnson, the creator of Spring, started exploring that space. [... more about Josh. Bootiful Podcast]
Vaibhav Gupta leads Boundary (YC w23), he hates ugly code. And as we know, most agent systems out there wouldn't pass a code review. So Viabhav is building BAML for clean, reliable AI infra. [AI that works podcast]
Isaac Miller is a long-time core contributor for DSPy, working on typing, fine-tuning, RL infra, and multi-modal support.
Hugh McKee is a skilled developer advocate with decades of experience building enterprise applications. Hugh is an expert in distributed systems and Akka, and can help connect all the dots!
Arun Gupta is a globally recognized expert in developer relations and open source strategy. With over 25 years of experience at Intel, Apple, Amazon, and Red Hat, he’s built developer communities of thousands of people.
Claire Longo is a mathematician and AI leader currently serving as Lead AI Researcher at Comet. She’s led teams at Twilio, Opendoor, and Arize AI, mentors engineers, and actively champions diversity in AI.
Jeremy Adams is a Senior Developer Advocate at Neo4j who blends deep technical understanding with strategic insight. He’s passionate about people, entrepreneurship, integration, and automation, and has worked across technical and business roles at Dagger, GitHub, Twistlock, and Puppet.
Fred Patton is a Senior Developer Advocate at Auth0 for Okta. He helps teams design AI systems that are scalable, accountable, and identity-aware. Prior to Auth0, he led developer-facing work on streaming data platforms at Nstream (formerly Swim.ai), working in security and healthtech.
Prashanth Rao is an AI engineer at LanceDB. He's spent a lot of time working on use cases related to RAG, knowledge graphs, vector search, and incorporating them as tools for agents using modern frameworks like DSPy and BAML. He writes about these techniques on his blog, thedataquarry.com.
More to be announced soon
Mentors are also familiar with MCP which you may want to use for data/process integration.
Please keep in mind, the mentoring and technical support are available but limited. Your team should have all the skills to troubleshoot any technical issues otherwise.
RUBRIC
Projects will be evaluated on production readiness, innovation, technical complexity, practicality, user experience, and overall quality. A detailed judging rubric will be provided to participants closer to the hackathon date.
AWARDS
Top 5 winners will have an opportunity to demo their projects on the main stage of the AI By the Bay conference.
🔥 Prize: $2000 1st, $1000 2nd, $1000 3rd. Top teams give lightning talks on the main stage.
USEFUL RESOURCES TO GET STARTED
Spring AI Project, Docs, Short Intro, Samples on GitHub
Josh's talks on Spring AI Session with James Session at SFJava
Rod Johnson's take Docs, Embabel Agent on GitHub, Blog
Amazon Bedrock Docs
Langchain4j Docs, Repo of Examples
DSPy Video
RAG in DSPy, Optimizing a program for math reasoning, Building an Agent with DSPy
Omar Khattab, DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines
Akka, What is Agentic AI?, Agentic tutorials and AI app samples
SUPPORT US EVEN MORE!
Interested in partnering? Email [email protected]