Building products with AI agents
Operating models, gates and evals that let one person direct a team of agents and ship working services.
Case studies
Three ChatGPT plugins that give the assistant ready workflows and tools: for developing Hintapiste.fi, for SEO and UX work, and for comparing purchases.
Price comparison built solo with AI agents.
Agent for a sales and marketing team: answers from an agreed knowledge base and asks for approval before changing work systems.
Tool for developing a large healthcare website: classifies search data automatically and shows the sources.
MCP server that lets ChatGPT code on a local Mac in allowed projects only. Every action is logged.
Tools for steering software work done by agents. Every step leaves a result you can check.
Operating model for a solo founder building with AI agents: phase gates, evidence classes and maturity states, with templates and a validator.
Guides, methods and benchmarks
Running agent work through gates
A guide to running coding agents through a scheduler, a gate and a main check, so that one person still decides what goes live.
Governance rules for coding agents
A guide to a rulebook for coding agents: one entry file, a reading order, safety limits, a task for every change and a check that runs before the merge.
A guide to an eval suite for agent output: one decision boundary per case, self-contained prompts, adaptive repetition, changes on repeated evidence only.
A guide to demos with no server: one export per snapshot, allowlisted fields, a closed schema, a redaction scan and a page that works without the file.
Working method and demos
Two days of Hintapiste.fi’s development from its event log: agent runs by lane, checks and releases.
How Hintapiste.fi works, shown on 5 products from a snapshot of its public data: price history, listings joined into one product and the order of offers.