Full‑Stack AI Engineer

Company Mission

Payhawk is a leading global spend management solution for scaling businesses. Headquartered in London and combining company cards, reimbursable expenses and accounts payable into a single product; its future-facing technology enables finance teams to control and automate company spending at scale.

The Payhawk customer base includes fast-growing and mature multinational companies in 32 countries including LuxAir and Wagestream. With offices in New York, London, Berlin, Munich , Barcelona, Paris, Amsterdam, Vilnius and Sofia; Payhawk is backed by renowned investors such as Lightspeed Venture Partners, Greenoaks, QED Investors, Bek Ventures and Eleven Ventures.

Our values include supporting flat hierarchies, taking ownership and responsibility, seeking and providing feedback, managing constructive critique, and speaking our minds. We understand that the best ideas don’t all come from the same place, so we encourage diversity and inclusion in all areas of our work.

The future of fintech is about more than money, and we believe in work-life balance, continual learning, and empowered teams. We’re also on a journey to measure and improve our environmental and social impact. From virtual cards to digital subscriptions, our software and automation help take paper out of the equation for our customers, too.

We’re changing the world of payments, and we’re looking for an exceptional team to help us. 

 

About the Role

Automation, Integrations, Agents, Platform, Retrieval & AI UX

As part of our Efficiency team, you will upskill and uplift every role with AI‑powered tools - automating workflows and building secure, reliable assistants, agents, and role “digital twins.” Just like broader Payhawk we obsess over customer success, our customers are every Hawk and every Payhawk customer, and we measure impact by adoption, task success, and time saved. We design human‑centered AI and manage risk responsibly.

You’re a product‑minded software engineer who designs and ships end‑to‑end AI solutions: chat UIs, APIs, orchestration, data pipelines, retrieval, evals, and guardrails. You love jumping between frontend, backend, prompts, schemas, observability, and Kubernetes to deliver measurable time savings for Hawks.

Rigorous studies show AI assistance can raise knowledge‑worker productivity, especially for less‑experienced workers - so we measure against hours saved and task success from day one.

How we work

• We behave like a Platform + Enabling capability that accelerates stream‑aligned teams.
• We measure delivery with DORA’s Four Keys and product value with adoption & time saved. See why: DORA 2024 report.
• We build tools‑using agents via MCP and rely on RAG / Hybrid RAG for grounded answers.

 

Responsibilities

• Build UI + backend for chat/agent experiences; own discovery → shipping → monitoring.
• Orchestrate tool‑using agents via MCP; design safe tool APIs and connectors.
• Implement and iterate RAG / Hybrid RAG (dense + BM25) with quality evals (faithfulness, groundedness, latency); manage prompt & retrieval versioning.
• Data work: build ingestion/ETL, chunking/metadata, access‑aware retrieval; ensure lineage and governance.
• Reliability: tracing, cost & latency SLOs, autoscaling; CI/CD with quality gates (offline/online LLM evals).
• UX & prompts: craft conversational flows using People + AI Guidebook patterns; instrument analytics and A/Bs for task success and satisfaction.
• Security & risk: apply applicable security standards such as (but not limited to) NIST AI RMF guardrails; work with Security/Legal on data handling and auditability.

Requirements

• 4+ years experience; expert in Node.js/TypeScript/React/PostgreSQL; CI/CD & cloud (GCP/K8s).
• Shipped LLM apps (agents, function/tool use, retrieval); strong API design & data modeling.
• Comfortable with Python for data/ETL; vector stores; analytics instrumentation.
• Bachelor's degree in Computer Science (or related field).
• Experience with agile processes, continuous delivery, and continuous integration.
• Ability to multitask, organize, and prioritize work.
• Exceptional integrity and work ethic.

Nice to have

• Frontend experience, Kafka, experimentation platforms, prior experience with MCP servers/clients.

Company Benefits

• Competitive compensation package based on experience.
• 30 days holiday paid leave.
• One week exchange policy to another Payhawk office (London, Berlin, Barcelona, Paris, Amsterdam and Vilnius).
• Flexible working hours and opportunity to work from home.
• Regular team-wide events.
• Additional medical care.
• MultiSport card fully funded by us.
• Company office massages.
• Personal assistant service.
• Opportunity to use the Payhawk product (that is, essentially, built by you).