aggregate-intellect logo

Agentic BuildcampA cognitive gym for building AI agents — using AI agents.

You don't get stronger by watching someone lift. This is where pragmatic builders — developers, founders, and operators — ship a real AI agent in 7 weeks by training the way professionals do: structured reps, a cohort pushing alongside you, and coaches who fix your form before bad habits set in.

Amir Feizpour

Taught by Dr. Amir Feizpour — founder of Aggregate Intellect, ex-Oxford & RBC NLP Lead, a decade building production AI systems. Meet the mentors →

Is This For You?

Follow the path that describes how you want to build.

Agentic Buildcamp idea maze — find your path from learning goals to the right program

You'll Thrive Here If...

The strongest outcomes come from people who like building in public, iterating fast, and using the cohort to finish something real.

Committed Builders

You want to stop collecting courses and start finishing substantial work. This buildcamp rewards consistent effort and steady progress on a real project.

Self-Directed Problem Solvers

You like researching, experimenting, and figuring things out as you build. This is not a step-by-step course, so curiosity and initiative matter.

First-Principles Builders

You want to understand how agent systems work and build from first principles instead of relying on plug-and-play templates for someone else's problem.

Outcome-Focused Tool Mixers

You care more about shipping the right solution than mastering a single framework. We use whatever combination of tools fits the project best.

Portfolio-First Professionals

You care more about real applications, shipped demos, and proof of work than collecting certificates or adding another credential line to LinkedIn.

Rapid Iterators

You can start before every detail is mapped out. We move quickly, learn by doing, and improve the project through short feedback loops.

Experience level: You need to be willing to build using whatever tools work best - whether that's coding, AI-assisted development, no-code platforms, or combinations of all three. We support any approach that gets your project built. The main requirement is showing up ready to build.

How You Build

The program is delivered using Sherpa-B — an MCP server that acts as your building coach or thought partner. Concretely, this is a series of workflows you run directly in your agentic coding environment (eg. Claude Code). You learn the methodology by using it in action while building. Sherpa-B is built on KnowledgeOps — our framework for AI-enabled knowledge work.

Claude Code
See how the program is structured →

We have helped 250+ builders like you who've come from all types of industries and they have gotten great value from our bootcamp.

Where our alumni work

Don't take our word for it,

see what our alumni say

I love the bootcamp. Amir was very keen on refining the sticking points in the workflow that the AI Agents can perform. The mentors were very knowledgeable. We completed the main functionalities defined for our three projects. We are now continuing to expand the projects. I have learned a great deal from the bootcamp.

Alice R.

Postdoctoral Fellow, St. Michael Hospital

This bootcamp provided an exceptional learning experience that enhanced my practical understanding of AI agents and the very real considerations for their deployment across a variety business contexts. It fostered my ability to engage effectively with my clients in the financial services sector, enhancing my ability to explore potential use cases with them.

Fredrick F.

Principal, Microsoft

I enjoyed the level-headed discussions around the multi-agent frameworks and liked the guidance we got when building our own product. The demos exceeded our own expectations. It is exciting to have my team interested in continuing to work on the project beyond the course. We plan to continue A lot of effort in making project demos.

Maher L.

CTO

It was an intense and rewarding 7-week journey to build a real product from ideation to a fully functional application. With the support of the instructors, I challenged myself to create a multi-agent system, and the experience felt remarkably close to working in a real software company; delivering features sprint by sprint. The instructors were highly responsive, and the course structure was well thought out.

Sheikh A.

AI Developer Advocate

I have to say, this bootcamp was a game-changer! As someone who isn't a native coder, I was genuinely surprised and thrilled that I was able to build a full-fledged agentic application. What really sets this apart is how incredibly practical it is, a refreshing change from a lot of the theory-heavy courses out there! I came in with just a basic idea, but the discussions and guidance really expanded my thought process.

Sorabh V.

Principal PM Microsoft

Although I have a background in computer science, building an agentic system was new to me. I came in with an idea but was concerned about pulling it off. The playbook with guided steps for refining ideas, the sample project, supporting resources, regular follow-ups, and 1-on-1 meetings with knowledgeable mentors helped me stay focused, complete my project, and gain confidence in abilities.

Nasim S.

Machine Learning Engineer

Featured Projects

These are our favorite projects from the recent bootcamp cohorts.

Featured projectPlay
Featured projectPlay
Featured projectPlay

What You'll Actually Get

This isn't just another course. You'll walk away with tangible assets that advance your career and concrete skills that set you apart.

A Complete AI Agent Application

Build and deploy a fully functional agentic AI application that solves a real workflow problem. Not a toy project — a showcase-ready application you can demo to employers, clients, or investors.

Public Demo & Portfolio Piece

Present your project to hundreds of attendees in our public demo showcase. Get a professional demo video and case study that proves your AI engineering capabilities.

Real-World Complexity, Not Toy Demos

Build under conditions that reflect how agents actually behave in production — edge cases, failure modes, and the nuance that vendor demos skip.

Weekly 1:1 Sessions with Working AI Engineers ⭐

Meet regularly with mentors who build AI systems professionally. Use those sessions to unblock architecture decisions, review implementation tradeoffs, and tighten your demo before launch.

KnowledgeOps Methodology

The KnowledgeOps methodology — 12 guided workflows for building real-world AI agents. From problem framing and workflow design to evaluation and deployment. A process you internalize and take with you, independent of any single tool or framework.

Prompts and Workflows You Keep

The baseline prompts and workflow templates from the program are yours. Refined through the buildcamp, they are your starting point for every agentic project you build after.

⭐ = Mentorship Tier Only

Bottom line: You'll leave with a shipped agent, the KnowledgeOps methodology internalized, and a cohort of builders you actually worked with.

FAQs

What other costs should I expect beyond the program fee?

Plus icon

What can I build in this buildcamp?

Plus icon

What is the expected time commitment?

Plus icon

What happens if I start the cohort and then realize that can’t continue?

Plus icon

What’s the refund policy?

Plus icon
Aggregate Logo Icon