Tech Lead

We are looking for someone who does four things well. Designs systems and can defend the design. Leads a small team and makes them better. Communicates clearly enough that neither the engineers nor the client are ever guessing. And uses AI tooling with enough judgement that the team ships faster without quietly accumulating code nobody understands. The work is mainly Python and JavaScript, but we are not prescriptive about frameworks and we are not counting years against a specific tool. What matters is that you have taken real systems from design to production and can explain the trade offs you made. Depth in one stack and the appetite to work in another is the profile we want. The role is genuinely full-stack. You would be designing services and data models one week and building the interface on top of them the next, and we expect real ownership of both. The products are production systems in regulated and data-heavy domains, so correctness, auditability and uptime count for more than shipping quickly and revisiting later.

What You Will Do

• Own the architecture on a product or engagement. Service boundaries, data model, integration points, and the reasoning behind each one written down somewhere a new joiner can read it.
• Design for the change you already know is coming. Most of the cost in these systems lands on whoever extends them later.
• Make the non-functional requirements explicit early. Throughput, latency, failure modes, recovery, audit trail.
• These get decided at design time or they get discovered in production.
• Lead a small team day to day.
• Sequence the work so nobody is blocked, review code in a way that teaches, and make the call when the team is genuinely split. Grow the engineers around you.
• A lead who ships everything personally has not led anything. Set how the team uses AI tooling, including the review bar for code nobody typed by hand, and decide where a model belongs in the product itself and where a deterministic rule is the better answer.
• Work directly with the client and their product people. A lot of the value here is asking the question that stops three weeks of wrong work. Stay with what you build. Take it to production and own the pipelines that get it there.