LLM Engineer
Build production AI features inside our enterprise systems: retrieval-augmented assistants (RAG), document-aware chat, and structured extraction from unstructured input. You own the model layer end to end, from prompt design to evaluation and production deployment.
Requirements
- Hands-on experience with OpenAI, Anthropic, or open-weights models in production, not just research notebooks
- Comfortable with vector databases (pgvector, Pinecone, Weaviate) and building retrieval pipelines
- Understand Prompt Engineering as a real discipline: structured outputs, Function Calling, eval harnesses
- Strong Python. Comfortable building APIs that wrap model calls with caching, retries, and observability
- Have shipped at least one feature where you measured model quality with a real eval set, not by feel
- Pragmatic about cost and fit: you know when a large model is overkill, and when fine-tuning a smaller one is the better path
Responsibilities
- Design and build AI features inside our enterprise platforms: chat, extraction, summarization, classification
- Build retrieval pipelines: chunking, embedding, ranking, citation tracking
- Write eval harnesses so we can measure the impact of prompt changes objectively
- Monitor production model usage: latency, cost, failure modes, drift
- Pair with backend engineers to wire the model layer into our .NET services
Benefits
- Work on real production AI features engineered into enterprise platforms our team operates
- Fully remote across the Gulf and Egypt time zones, no mandatory office
- Full ownership of the model layer, from design to production monitoring
- Direct access to founders and senior engineers, no middle management between you and the work
- Continuous technical exposure to a multi-discipline team: Backend, DevOps, QA, and product
Apply
Send your CV or LinkedIn. We reply within 72 hours.