✓ ShippedByAgents
Practitioner (own blog/paper) New product

QuEra: one engineer with Cursor and Claude Code merges 111 PRs across 19 repos, empty repo to merged service in six days

QuEra, a quantum computing company, reports results from an effort where one engineer used Cursor and Claude Code in interactive sessions, while also working on other projects, to build two new services (task profiles and user groups). The review recorded 101 completed tasks and 111 merged pull requests across 19 repositories, adding about 11,200 lines of production code and 25,400 lines of tests. Task profiles went from an empty repository to a merged service in six days and ran on a testbed and a real QPU in seven; user groups took 88 days to reach all four clusters. AI costs were about $3,900 against an internal estimate of 9.5–13.5 engineer-months. Separately, QuEra built Quantopus, a coordinator for worker, reviewer, planner and nightly testing agents, whose first project was the Admin Usage Recorder.

The numbers (as reported)

Merged pull requests111 across 19 repositories
Completed tasks101
Code added~11,200 lines production, ~25,400 lines tests
Empty repo to merged service (task profiles)6 days (running on testbed and real QPU in 7)
User groups to all four clusters88 days, engineer also on other projects
AI cost~$3,900 across both projects
Internal estimate for same scope9.5–13.5 engineer-months ($144,000–$205,000 salary), ±50%
Failed reviewer run~500M cached-input tokens, ~$1,000

Facts

Company / subject
QuEra
Tools
Cursor, Claude Code
In-house tools
Quantopus (QuEra)
Codebase
greenfield
Work type
New product
Stack
—
Published
2026-11-10
Last checked
2026-11-11

What the source claims Practitioner (own blog/paper)

Claimed by the builder, writing about their own work. Not checked independently.

  • Task profiles went from an empty repository to a merged service in six days 6 days
    “Task profiles went from an empty repository to a merged service in six days”
  • 111 PRs merged across 19 repositories, adding about 11,200 lines of production code and 25,400 lines of tests 111 merged PRs
  • AI costs were about $3,900 against an estimated 9.5–13.5 engineer-months for the same scope ~$3,900 vs 9.5–13.5 engineer-months
    “That points to significant potential savings, but the comparison is incomplete.”

Caveats

Self-reported by QuEra. The 111-PR results come from interactive Cursor and Claude Code sessions by one engineer, not from unattended Quantopus runs, and the post keeps the two apart. QuEra itself says the $3,900 excludes the engineer's time and infrastructure, prices Claude at API rates rather than the actual subscription and misses earlier usage, that the effort estimate has about ±50% uncertainty, that there is no comparable project to measure speedup, and that test results were not independently rerun. All 111 PRs were merged by the engineer; 54 also had peer review. No author is named.

More stories by tool: Claude CodeCursor

Related topics: New products built with agentsGreenfield projects

Source

QuEra blog

One email a week: what actually shipped

New sourced stories, the numbers behind them, and what didn't work. No hype.

We'll only use this to send the newsletter. See the privacy page.