Mission 11 of 12 · Updated November 2026

Build Your AI-Native Capstone

Two weeks, your product. Scope a real AI-native application - a recruitment platform, customer-support assistant, sales research agent, productivity agent or document processor - then build it end to end: frontend, backend, database and one genuinely useful AI capability, with tools where they fit, tests and Langfuse traces. Extending your recruitment project is allowed if the new work is substantial and yours.

By the end of this mission: “I can scope and build a full-stack, AI-native application end to end, using coding agents responsibly.”

The tech stack for this mission

React logoReactVite logoViteExpress logoExpressPostgreSQL logoPostgreSQLOpenAI logoOpenAIZod logoZodMCP logoMCPLangfuse logoLangfuseCodex CLI logoCodex CLIGitHub logoGitHub

What you learn, checkpoint by checkpoint

  1. Checkpoint A

    Scope the product, plan the architecture and set up the skeleton

    • Scope the product and define the smallest version that is genuinely useful
    • Plan the architecture with clear separation of concerns
    • Set up the full-stack skeleton: React, Express, PostgreSQL

  2. Checkpoint B

    Build the core features, data model and one genuinely useful AI capability

    • Implement the core features and data model
    • Design and build one genuinely useful AI capability
    • Manage context and tools so the AI gets exactly what it needs

  3. Checkpoint C

    Add tools where they fit, error handling, tests and Langfuse traces

    • Integrate tools or MCP where the product needs them
    • Test and debug incrementally, with basic error handling throughout
    • Use coding agents responsibly - you review, you understand, you own the code

AI companion mode · AI as Operator

You direct the coding agent

You direct an AI coding agent through a full ship cycle. Approve plans, reject bad diffs, gate every deploy, own rollbacks.

What proves you’re done

Working capstone (frontend + backend + database + AI) with tests and Langfuse traces

Every mission ships something real you can show. If you can’t explain it, you can’t submit it.

Deliberately left out of this mission

Mandatory multi-agent architectures · Complex infrastructure · Additional frameworks

Depth is set on purpose: you master what this mission needs, not everything at once. The advanced material lives in optional self-study after Mission 12.

Free · no card needed

Reading is not building.

This mission is part of the paid 12-week programme - but the way we teach is free to try. The free trial puts you inside Sigmo, the learning portal our students use daily, with the 6 Projects in 6 Days precourse: real projects, real code, published to your own GitHub.

Start the free trial

Mission 10

Make the Agent Reliable: Tracing, Evals & Safety

Mission 12

Ship, Document & Defend Your Capstone

Ready for the real thing?

The AI-Native Software Development Programme - all 12 missions with a live cohort, daily Buildroom, and mentors watching your back.

Apply nowSee the whole curriculum