Daniel — Autonomous AI Software Engineer

Daniel is an autonomous AI software engineer built by Cognitive AI. It reads your codebase, writes code, runs tests, and opens pull requests — without you doing anything manually.

🔗 Try it: cognitivedaniel.com | 📊 Benchmark: Daniel vs Devin

Benchmark results (v7, October 2026)

Daniel was tested head-to-head against Devin on 6 real bugs from popular open-source projects. Both agents received identical prompts. Each PR was graded by the project's own hidden test suite.

Metric	Daniel	Devin
Pass rate	6/6 (100%)	6/6 (100%)
Total cost	$1.45	$4.91
Total time	12m 33s	12m 12s
Cost per resolved bug	$0.24	$0.82

Daniel is 70% cheaper than Devin at the same pass rate.

Dataset: cognitivedaniel/daniel-benchmark-v1

How Daniel works

Daniel uses a 4-turn agentic loop:

Analyze — reads the issue, the failing tests, and the relevant code
Implement — makes the minimal code change to fix the bug
Test — runs the project's existing test suite to verify the fix
PR — opens a pull request with a clear description

The loop runs on Anthropic's Claude API. Daniel never opens a PR it hasn't verified against your tests.

Supported languages

Python, JavaScript, TypeScript (primary)
Java, Go, Rust (via Multi-SWE-bench tier, in testing)

Integrations

GitHub (repos, issues, PRs)
Jira (coming soon)
Slack (coming soon)
Run locally on any folder with one command

Pricing

Plan	Price	Tasks/month
Starter	$20/mo	20
Pro	$100/mo	200
Enterprise	$500/mo	Unlimited

Citation

If you use Daniel's benchmark results in your research:

@misc{cognitive-ai-daniel-2026,
  title={Daniel: Autonomous AI Software Engineer — Public Benchmark v1},
  author={Cognitive AI},
  year={2026},
  url={https://www.cognitivedaniel.com/devin-alternative},
  note={Dataset: huggingface.co/datasets/cognitivedaniel/daniel-benchmark-v1}
}

Downloads last month: -; Downloads are not tracked for this model. How to track

cognitivedaniel
/

daniel