$ whoamiSoftware Engineer · AI at ZebracatAi, where we turn text into video with generative AI. I work across the full stack but live in the backend — designing distributed services and async pipelines that stay fast and predictable at high request volume.
I'm an AI enthusiast in the practical sense: I build and ship LLM features — agents, RAG, guardrails, and the cost engineering that keeps them production-grade. I also enjoy media-heavy work — programmatic video with tools like Remotion and headless rendering. I read a lot, write about what I learn, and I'm happiest when I'm shipping.
- 🧠 Production AI — LLM agents · RAG pipelines · guardrails · LLM cost engineering
- ⚙️ Backend at scale — Python · Django · FastAPI · async queues · microservices
- 🎨 Frontend — React · TypeScript · Tailwind · Framer Motion — fast, polished UIs
- 🎬 Media — programmatic video (Remotion) · headless rendering pipelines
- 🔭 Reliability — observability, caching & back-pressure (Sentry · Prometheus · Grafana)
- ✍️ I write about real-world AI systems on Medium
- Stop Treating LLM Calls as Free — How Real Cost Engineering for Agents Actually Works
- Stop Trusting the Embedding — How Real RAG Pipelines Actually Work
- Stop Trusting the Model — How Real LLM Guardrails Actually Work
- Stop Putting Everything in One Prompt — How Real Research Agents Actually Work
More at medium.com/@el_mohamad →
AI / LLM
+ more
