IN 12 DAYS

super{set} Hackathon: Making Agents Reliable in the Enterprise

SSsuper{set}
view event ↗
when4 Dec 2026
whereSan Francisco
priceFree
formatIn-Person
organisersuper{set}

about this event

December 4 | 9-Hour Hackathon | Hosted by super{set}

AI agents are becoming increasingly capable, but deploying them in the enterprise introduces a different set of challenges: how do you make them reliable, observable, secure, and scalable? super{set} runs a 9-hour hackathon focused on Making Agents Reliable in the Enterprise. Bring your ideas, build alongside fellow developers and AI enthusiasts, and tackle real-world challenges at the intersection of AI agents and enterprise infrastructure. The goal isn't just to build another AI demo; it's to create a functional, thoughtful product that addresses the challenges of bringing AI agents into real enterprise environments. From agent orchestration and observability to evaluation frameworks and privacy, participants explore what it takes to make AI agents work reliably in the real world.

The Theme: Making Agents Reliable in the Enterprise. Build solutions that make AI agents more trustworthy, measurable, secure, and production-ready. The organisers are particularly interested in solutions that explore: Observability: how can enterprises monitor, trace, and understand what their agents are doing? How do we identify failures and debug agent behavior? Evaluations (Evals): how can we measure agent performance, accuracy, and consistency? How do we know when an agent is ready for production? Orchestration: how can multiple agents, tools, and workflows work together reliably? How do we manage complex agent systems at scale? Privacy & Security: how can agents interact with sensitive information while maintaining privacy, security, and enterprise compliance?

Hackathon Tracks. Choose one of the following tracks to guide your build. Track 1: Reliable Agents for GTM & Marketing — build AI agents that make go-to-market and marketing workflows smarter, more autonomous, and more reliable. Think about agents that can automate lead research, prospecting, and enrichment using public data; coordinate multi-step GTM workflows across different tools and platforms; analyze marketing performance and generate actionable insights; evaluate their own outputs and ensure accuracy before taking action; provide visibility into agent decisions, actions, and performance. The challenge: how can we build GTM agents that businesses can actually trust to operate in real-world workflows? Track 2: Reliability in Privacy Agents — build AI agents that can work with sensitive or personal information while preserving privacy and maintaining trust. Think about agents that can work with public databases and APIs while respecting data privacy; identify and protect sensitive information; manage data access, permissions, and information flows; provide transparency into how data is accessed and used; ensure reliable outputs without compromising privacy. The challenge: how can we make AI agents useful and powerful without sacrificing privacy, security, or user trust?

What You Can Build With: public datasets and databases; public APIs; your favorite AI models, frameworks, and developer tools; agent orchestration, evaluation, and observability tools.

Judging Criteria. The organisers are looking for more than an interesting technical demo. Think of what you're building as a product: it should solve a meaningful problem, demonstrate technical depth, and have a realistic path to enterprise adoption. Projects will be evaluated on: Technical Depth: how thoughtfully and effectively does the solution use AI agents, orchestration, evaluations, observability, or other relevant technologies? Enterprise Readiness: how realistic is the solution for deployment in an actual enterprise environment? Does it address reliability, security, scalability, and practical implementation? Product Thinking: is there a clearly defined problem and target user? Does the solution deliver meaningful value, and could it evolve into a real product?

Prizes. Winning teams will receive an opportunity to take their projects and careers to the next level. Each winning team will receive one of the following opportunities: an interview with super{set} or one of its portfolio companies; a VC pitch workshopping session with a super{set} General Partner.

Event Agenda, December 4 | 9-Hour Hackathon — a full day of building, collaboration, mentorship, and demos. 10:00 AM – 10:30 AM: Check-in & Networking; 10:30 AM – 11:00 AM: Welcome, Introductions & Hackathon Kickoff; 11:00 AM – 5:00 PM: Hacking, Building (with lunch, coffee, and snacks provided); 4:50 PM: Project Submissions DUE; 5:00 PM – 6:00 PM: Project Demos; 6:00 PM: Judge Deliberation; 6:30 PM: Winners Announced + Closing Ceremony.

Who Should Attend? Software engineers and developers; AI/ML engineers and researchers; founders and aspiring entrepreneurs; product builders and designers; anyone excited about building the next generation of reliable AI systems.

tracks

Reliable Agents for GTM & Marketing

Build AI agents that make go-to-market and marketing workflows smarter, more autonomous, and more reliable, such as agents that automate lead research, prospecting, and enrichment using public data, coordinate multi-step GTM workflows across different tools and platforms, analyze marketing performance and generate actionable insights, evaluate their own outputs and ensure accuracy before taking action, and provide visibility into agent decisions, actions, and performance.

Reliability in Privacy Agents

Build AI agents that can work with sensitive or personal information while preserving privacy and maintaining trust, such as agents that work with public databases and APIs while respecting data privacy, identify and protect sensitive information, manage data access, permissions, and information flows, provide transparency into how data is accessed and used, and ensure reliable outputs without compromising privacy.

when

4 December 202610:00-10:30 PST

Check-in & Networking

Check-in & Networking.

◍ San Francisco, CA
4 December 202610:30-11:00 PST

Welcome, Introductions & Hackathon Kickoff

Welcome, Introductions & Hackathon Kickoff.

◍ San Francisco, CA
4 December 202611:00-17:00 PST

Hacking, Building

Hacking, Building, with lunch, coffee, and snacks provided.

◍ San Francisco, CA
4 December 202616:50 PST

Project Submissions DUE

Project Submissions DUE.

◍ San Francisco, CA
4 December 202617:00-18:00 PST

Project Demos

Project Demos.

◍ San Francisco, CA
4 December 202618:00 PST

Judge Deliberation

Judge Deliberation.

◍ San Francisco, CA
4 December 202618:30 PST

Winners Announced + Closing Ceremony

Winners Announced + Closing Ceremony.

◍ San Francisco, CA

tickets

Ticket types, prices and availability for super{set} Hackathon: Making Agents Reliable in the Enterprise
ticketspriceFeeStatusdescription
StandardFreeFreeAvailableRegistration requires host approval.

prizes

Prize places and values for super{set} Hackathon: Making Agents Reliable in the Enterprise
#PRIZEprice
Winning teamsAn interview with super{set} or one of its portfolio companies, or a VC pitch workshopping session with a super{set} General PartnerTBA

faq

Does it cost anything to attend?+
The Standard ticket is free. Registration requires host approval, so your place is confirmed once it is approved.
How do I register?+
Register through https://luma.com/superset-6lth. Registration asks for your LinkedIn profile, which best describes you (undergrad student in the third year or above, master's/PhD student, working engineer or developer, founder, product or design, or other), what you built and your role, a description of a time something you built broke or behaved unexpectedly, and optionally a link to what you built such as GitHub, a portfolio, a demo or a write-up.
Who should attend?+
Software engineers and developers; AI/ML engineers and researchers; founders and aspiring entrepreneurs; product builders and designers; and anyone excited about building the next generation of reliable AI systems.
What can I build with?+
Public datasets and databases, public APIs, your favorite AI models, frameworks, and developer tools, and agent orchestration, evaluation, and observability tools.
What happens during the day?+
Check-in and networking run 10:00 AM to 10:30 AM, the welcome, introductions and kickoff 10:30 AM to 11:00 AM, hacking and building 11:00 AM to 5:00 PM with lunch, coffee, and snacks provided, project submissions are due at 4:50 PM, project demos run 5:00 PM to 6:00 PM, judge deliberation is at 6:00 PM, and winners are announced with the closing ceremony at 6:30 PM.
super{set} Hackathon: Making Agents Reliable in the Enterprise4 Dec 2026 · San Francisco
view event →

never miss a hackathon

one email a week with every new listing. no spam, unsubscribe any time.