About the role
We're hiring an SDE 2/3 to lead an enterprise project building agentic workflows. You'll work closely with the client to understand their current processes, figure out how to agentify them, finalise the flows, and then lead the team to build them — while doing hands-on IC work yourself. You'll be the client's point of contact for everything on the project.
Responsibilities:
- Work with the client to understand their current process and identify what can be agentified.
- Discuss and propose agentic flows with the client, and finalise them together.
- Design the agent architecture and overall system for the product.
- Lead a small team of engineers while contributing as a hands-on IC.
- Act as the client's single point of contact for scope, timelines, demos, and escalations.
- Bring product thinking: prioritise what to build first and what to leave out.
- Has led client conversations — requirement discussions, demos, and stakeholder management.
- Has led engineers to deliver projects on time — planning, code and design reviews.
- Can handle difficult situations — escalations, scope changes, tight deadlines, and tough client conversations.
- Own end-to-end project delivery — scope, timelines, quality, and outcomes.
The core requirements for the job include the following:
Must have:
- Strong Python, with hands-on experience in frameworks like FastAPI and Django.
- Databases: SQL and NoSQL, schema design, query optimisation.
- Caching: Redis or similar.
- System design: APIs, queues, scalability, failure handling.
- Agent loops: tool calling, state, retries, stopping conditions.
- Agent architecture: single vs multi-agent, orchestration, memory, human-in-the-loop.
- LLM engineering: prompting, RAG, structured outputs, evals.
- Product thinking.
- Ownership: accountable for project delivery end-to-end.
Good to have:
- Agent frameworks (LangGraph, Claude Agent SDK, OpenAI Agents SDK).
- Cloud deployment (Azure, AWS, or GCP), Docker, CI/CD.