About the role
This role owns the architecture of production agentic systems in Python, from the first whiteboard discussion through deployment and long-term operation.
Technical architecture is only part of the role. You'll translate ambiguous business problems into systems that engineers can build, operations can support, and clients can trust. You'll set the technical direction across multiple engagements while remaining hands-on in design, implementation, and engineering reviews.
What you'll do
- Own agentic system architecture end to end: tools, orchestration, state, guardrails, human-in-the-loop.
- Lead solution architecture with clients and defend the trade-offs in writing.
- Build production-grade Python: typed, tested, observable.
- Design and maintain the eval harness for every system we ship.
- Own retrieval where it earns its place: RAG, reranking, grounding.
- Instrument cost, latency, and tool-call outcomes, and act on them.
- Set technical direction across two to four engagements and hold the bar in review.
Must have
- Around 6+ years of professional Python engineering experience, with a strong track record of delivering production systems, not notebooks.
- Agentic framework experience (required): you've shipped and operated multi-step, tool-using agents.
- Solution architecture (required): you own the design conversation with a client.
- LLM systems in production: prompting, context strategy, tool schemas, structured output.
- Eval discipline: you can show how you measured an agent.
- Strong API and backend design; async Python and queues hold no surprises.
- Comfort owning the path to production: containers, CI, deploys, observability.
- Excellent written architecture: the design doc is the deliverable.
Nice to have
- Deep experience designing Model Context Protocol (MCP) servers.
- Experience building Retrieval-Augmented Generation (RAG) systems at production scale, including retrieval evaluation.
- Experience with fine-tuning or model distillation, or a strong understanding of when neither is the right solution.
- Data engineering background.
- Experience delivering solutions for regulated industries such as government, financial services, or healthcare.
How you work
We value these qualities as highly as your technical expertise.
- You enjoy solving complex engineering problems without making simple problems unnecessarily complex.
- You write code and architecture that other engineers can understand, maintain, and extend.
- You communicate technical decisions clearly, whether you're speaking with engineers, project managers, or executive stakeholders.
- You measure system quality rather than relying on intuition, and you expect decisions to be backed by evidence.
- You mentor other engineers through thoughtful reviews, coaching, and technical leadership.
- You take ownership of systems after deployment and continuously improve reliability, performance, and operational quality.
- You raise risks early and communicate trade-offs honestly before they become production issues.
- You work at a sustainable pace and help build engineering practices that scale with the team.
- Professional working proficiency in English and Arabic. You'll communicate directly with client stakeholders in both languages, so both are required for this role.
Stack
- Python
- MCP
- FastAPI
- Postgres + pgvector
- Docker
- Kubernetes
- AWS