Senior AI Engineer (Python / LangChain / LangGraph)

Company: warpSpeed (UK)

Location: Fully remote, India (with a daily overlap with UK working hours)

Job type: Full-time

Experience: 6+ years in software engineering, including 2+ years building LLM-powered applications in production

Salary: Highly competitive

About warpSpeed:

warpSpeed is building the world’s most personalised AI assistant. Our mobile app brings a person’s email, calendar, notes, tasks and messaging together across all their accounts, Gmail, Outlook and more and puts an AI assistant on top that understands how they work and helps them stay on top of everything.

We’re not building another inbox or another to-do list. We’re building one place where the AI can see the whole picture and act on it: triaging email, scheduling, drafting, following up and turning conversations into work that gets done.

Who you’ll be working for:

warpSpeed is a UK-based company founded by Martin Warner, a serial entrepreneur with a track record of building and scaling technology ventures. It’s a founder-led team where engineering is a first-class priority: your work is visible, valued and central to every release.

About this role:

This is a full-time remote role for candidates based in India. You’ll work directly with our UK-based leadership, product, and marketing teams on a product used by a global audience. Compensation is competitive within the Indian market and will be determined by your experience and expertise.

We’re looking for a senior AI engineer to own the warpSpeed AI assistant and lead a small team of two to three engineers working on it. You’ll report to the CTO, but day to day you’ll run AI engineering yourself: deciding how the assistant should reason, retrieve and act, making architecture calls, reviewing code and prompts, and proving with evaluations that changes actually make it better. The assistant is a multi-agent system in production. A LangGraph supervisor agent breaks down each user request and routes it to specialised sub-agents for email, calendar, tasks, notes, messaging, projects, web search, the product knowledge base and more. Those agents retrieve the user’s own data from vector indexes, call back into our platform to take actions, and return answers with references and suggested next steps. Your job is to make that system more capable, more reliable, faster and cheaper.

This is a hands-on engineering role, not a research role. We care about shipping AI features real people rely on every day, and about the software engineering that makes that possible.

What you’ll do:

Lead AI engineering

• Take product requirements and turn them into agent designs, prompts, tools and well-scoped work for the team.
• Set and uphold engineering standards for AI code: testing, evaluation, logging and prompt management.
• Review pull requests and prompt changes, and mentor and unblock the other engineers.
• Own delivery: estimate honestly, flag risks early and keep the CTO informed without needing to be chased.

Own the assistant

• Design, build and improve the supervisor and sub-agents: routing, tool design, planning, memory and multi-step actions.
• Improve retrieval over users’ email, calendar, tasks, notes and messages: chunking, embeddings, hybrid and filtered search, ranking and context construction.
• Own the evaluation process. Grow our regression evaluation suites, measure against a real noise floor and make sure prompt and model changes are shown to help before they ship.
• Manage latency and cost: model selection, caching, token budgets and rate limiting. Every token we use turns into credits a user pays for, so accurate usage accounting matters.
• Build personalisation: help the assistant learn how each user works and adapt to it.
• Make the assistant safe to trust with someone’s inbox. Handle untrusted content (such as prompt injection hidden in incoming email), protect personal data, and make sure actions are taken only when the user intends them.
• Own production quality: tracing, observability, error handling and incident follow-up for the AI service.

Work across the product

• Work closely with the backend team on the contract between our Node.js platform and the AI service: what data the assistant sees, which actions it can take and how results come back.
• Work with product and design to define what the assistant should do, and be honest about what current models can and can’t do reliably.
• Keep up with a fast-moving…