Mike

Hello 👋🏻

I build developer-first data infrastructure for turning messy, real-world content into structured inputs that AI systems can actually use.

My work focuses on the unglamorous but critical layer between raw data and intelligent behavior:
crawling, extraction, normalization, chunking, enrichment, and delivery — done thoughtfully and at scale.

---

🧠 What I care about

• Clean inputs over clever prompts
• Context over raw text
• Systems that are boring, predictable, and reliable
• Tools that remove weeks of manual work from AI pipelines

If an AI system behaves badly, the problem is usually upstream. That’s the part I like fixing.

---

🛠 What I build

• End-to-end Apify Actors (not just scrapers)
• Documentation → RAG / fine-tuning pipelines
• Semantic chunking and context preparation
• AI-ready datasets and ingestion layers
• Infrastructure that scales without surprises

Most of my work lives in the “plumbing” of modern AI systems — intentionally.

---

🚀 Featured Actor

• Context Layer
  Transforms documentation sites into structured, context-rich data optimized for RAG, fine-tuning, and semantic search.

---

⚒️ Tools & stack

• Apify platform & Actors
• Node.js / TypeScript
• HTML parsing & DOM normalization
• Token-aware chunking strategies
• OpenAI & Anthropic APIs
• Vector database–ready outputs

---

🫱🏻‍🫲🏽 Working style

I prefer building products, not one-off scripts.
Clear scope, honest pricing, predictable behavior.

If you’re using my Actors, the goal is simple:
they should feel like infrastructure you don’t have to think about.

---