We live in the age of AI-assisted software development. 

The amount of AI-generated code and technical debt has grown exponentially.

In 2024 alone, developers generated over 256 billion lines of AI-written code, a figure on track to double in 2025.

The consequence of this shift is that there will be even more non-human-validated code referenced by LLMs.

Have you noticed that a completely new job title has already emerged: the “vibe coding cleanup specialist”? Put that into LinkedIn People Search and see for yourself. 😅

Today, software engineers lose ~35–50% of their time chasing answers buried across multiple sources instead of writing code.

The need for finding up-to-date, context-specific code and documentation is bigger than ever before.

• LLMs hallucinate and surface outdated code
• Stack Overflow is dying a slow death
• GitHub’s search returns a long list of raw search results 

That is why we are building GitHits

GitHits is a code search engine for agents that distills ready-to-use, up-to-date code examples from all public GitHub repositories so you can go from a problem to a working snippet in minutes, not hours.

Our vision is to organize the world’s code and make it instantly accessible and useful.

Our goal is to build the fastest, most comprehensive code search engine – indexing all public repositories on GitHub, integrating with private repositories on GitHub, GitLab, and Bitbucket, and searching the open web – to power agents, IDEs, LLMs, and AI coding assistants like Cursor, Claude Code, and Lovable.

What makes GitHits different?

• No LLM hallucinations, but real, up-to-date code from GitHub, linked to live repos
• Not a list of raw search results, but distilled code examples
• Unlike repo-scoped tools, GitHits searches all public repositories on GitHub
• No keyword guesswork – describe your issue and GitHits’ agentic search does the rest

Other tools hand you barrels of grain, yeast, and water. GitHits distills a single malt for you. 🥃

Status

We’re currently in private beta, getting ready to launch our MCP server.

We haven’t even incorporated the company yet, so this is an ideal opportunity for an ambitious person to become a founding engineer, as we already have strong investor interest lined up.

The ideal candidate

We’re looking for a product engineer to help us go from 0→1.

• Has experience in scaling a developer tool
• Great UX/UI skills
• Has an internationally credible background
• Willing and able to go through Y Combinator
• Has already built a developer audience (on X/Twitter and/or via a newsletter)
• Has experience raising funding from VCs

You'd work closely with Olli-Pekka Heinisuo, the developer of the OpenCV-python package with 100M+ downloads.

If you tick the majority of the boxes or if you’re confident you’d kill in this role given the opportunity, we’d love to hear from you.