Skip to content
Back to all notes
Web & App Build6 min read

Devesh JoshiCo-founder, product

Nine years building AI platforms serving 12,000+ engineers. LLM platforms, agentic systems (MCP), and multi-model safety evaluation.

Integrating AI pull request reviews without introducing security vulnerabilities

Automating initial PR sanity checks, lint compliance, and security static analysis using LLMs embedded into GitHub Actions.

Preventing reviewer fatigue in rapid development

As engineering velocity increases, senior engineers spend hours reviewing boilerplate PRs. AI code reviewers flag missing error handles and unescaped queries before senior review.

Topic Focus & Target Concepts

AI pull request reviewautomated code review GitHub ActionsLLM code security scannerdevops AI automationagile engineering velocity

This is the work behind our build practice — pitch-ready MVPs through to production applications, on a stack chosen for whoever maintains it after we leave.

Product and web app builds

Running into this in your own stack? Twenty minutes, no deck.

Book the call