AI posts
Posts tagged "AI".
I Built an AI Security Auditor, Then Audited It. It Was Wrong Twice.
Six agents read what three scanners found and grade a container image. Then I checked every finding against docker inspect and a raw Trivy run - and found one recommendation that would break your image and one that was three releases stale. Both are in the README.
I Gave an AI Agent Write Access to My Codebase. Then I Took It Away.
AZ-CLAW is a terminal-first coding agent that never writes to disk. Every mutation lands in an in-memory overlay, you review a diff, and only approved actions are applied - plus a Groq to OpenRouter fallback that lives inside the model object.
AI Fills the Inventory. A Pure Function Picks the Dress.
VivahStyle is an in-store platform for Indian wedding boutiques - staff-operated, never customer-facing. Groq Vision writes the inventory and gpt-image-2 renders the try-on, but the recommendations are a deterministic scoring engine with no LLM anywhere near them.
I Got Distracted Watching a Golang Tutorial. So I Built a Project That Won't Let You.
LingoLearn turns any YouTube video into a quiz-driven lesson in 130+ languages. Built in seven days for the Lingo.dev hackathon - here's the architecture, the Next.js bug that ate an evening, and what I got wrong.
"Whatever you lose, you'll find it again. But what you throw away you'll never get back."
- Himura Kenshin, Rurouni Kenshin