- Role
- Integrator and lead pitcher
- Team
- 4
- For
- Enterprise briefs in retail, F&B, finance and mobility
- When
- 8 to 12 Jul 2026, Ho Chi Minh City
- Status
- Winners
In short.
Agentic AI Build Week set enterprise briefs across retail, food and beverage, finance and mobility. The first hours went into an agent management architecture, which then spun up hundreds of agents to build seven entries in five days.
Two entries won their tracks, Retail and Mobility, and two more were finalists. Every figure on stage traced back to evidence or was cut.
Team Truthworks split the work: reverse-engineered APIs, harnesses for the final agent harnesses, the visuals, and the backend infrastructure.
Coordination
- git
- append-only thread files
- claim board
Entries
- MCP servers
- Langfuse
- GLM-5
- React
- PWA
- Chrome extension
- LangChain
- Claude API
What this means for your business.
Many POCs at once
Instead of betting on one idea, agents build several proofs of concept in parallel, so you compare real results before choosing.
Test every assumption
Each assumption behind a project gets its own build and its own test, so weak ideas drop out early and cheaply.
Evals for each idea
Every POC gets individual evals that measure it against your goals, so the winner is picked on evidence, not on the best demo.
All projects.
Get in touch.
Have a question or a project in mind? Send me a message.