Four agents built their database in two weeks. They read zero lines
A MongoDB-like database for a hobby project, written spec-first by Claude, Kiro, Qwen Coder and Cursor. They read zero lines of the code and checked it through tests. The first version took about two weeks, part time.
Spotted: found in the wild. Details are as reported by the source above and have not been independently verified.
What they did
In August 2026 they started a hobby project building a MongoDB-like database, following a disciplined, spec-driven development model using Claude, Kiro, Qwen Coder, and Cursor. They read zero lines of the code the agents wrote, caring instead about correctness verified through tests and steering the product along the right roadmap. The first version was built in about two weeks of part time work.
How it went
From there they kept exploring: learning relational algebra, researching almost every kind of database, and reworking the internals with a query planner and executor. They say they learned more in those two months than in their previous 20 years in the industry, though the project isn’t open sourced yet.
Try it yourself with Claude
Pick a system you want to deeply understand, like [a database engine]. Write a full specification first, then have [Claude, Cursor, or another coding agent] implement every line from that spec while you focus only on correctness, tests, and the roadmap, without reading the generated code yourself.
A starting prompt written by Agents in the Wild. Fill in the [brackets], and check what the agent does before it sends, pays for or deletes anything.
After $45 in GitHub Actions overage in one month, a solo founder put an idle homelab to work as CI runners, with Claude making it easy to set up. Daily spend dropped from $2-4 a day to under $0.25.
A first homemade synth has 36 hall-effect keys and buttons and 7 rotary encoders. The firmware started by hand, then gentle prodding at Claude got everything needed much faster. Rev 2 of the PCB is nearly done.
Working on a Quake project, Claude decided on its own to build a test suite: demo playbacks with snapshots compared pixel by pixel, plus recorded demos whose screenshots it checked for each element.
A data scientist tried Cursor on a real project. Prototyping, code completion and commenting went well, but pandas code kept failing and it deleted a bunch of data files.
After building a minimal proof assistant, the writer asked Claude how they knew the prover was actually sound. Claude checked, found five different ways it could prove false theorems, and fixed them.
Four years after shelving a homemade RSS reader and selling off its $500 domain, the writer built it this year with Codex and Claude. They use it daily, along with 30 other strangers who found it public.