Field notes from production.

Case studies on the systems I've shipped — RAG pipelines, real-time LLM orchestration, and the unglamorous backend engineering that turns a prototype into a product.

• Building a Production RAG System: Fitmeal Planner's Nutrition Engine

  How I built a production RAG system that lets users modify a 7-day meal plan in plain English while enforcing hard allergy and calorie constraints — architecture, trade-offs, and what I'd do again.

  • production RAG system
  • RAG developer
  • AI engineer
  • LangChain

• Low-Latency LLM Orchestration: Real-Time Adversarial Personas in Hard Talk

  How I engineered Hard Talk — an AI rehearsal platform where adversarial LLM personas interrupt and challenge users in real time over WebSockets. The latency problems nobody warns you about, and how I solved them.

  • LLM orchestration
  • real-time AI
  • WebSockets
  • low latency LLM

• Designing a SaaS API Gateway with Tiered Access and Stripe Credit Billing

  How I architected Inflectiv — a scalable API gateway with custom rate limiting, auto-generated OpenAPI docs, multi-tier access control, and end-to-end Stripe credit billing. The monetization engineering behind a real SaaS product.

  • SaaS API gateway
  • Stripe billing
  • rate limiting
  • FastAPI