Backend Engineer (AI Platform) - London
£80k–£100k + EMI equity
About the RoleWe’re hiring a Backend Engineer to build the core infrastructure behind Revax’s AI-native platform for professional services.
You’ll design, optimise, and scale the systems that power multi-agent drafting, knowledge graphs, and auditable AI workflows used by accountants, tax advisors, and consultants at leading firms. This is not a “ticket-taking” role -you’ll work directly with the founders, own critical backend services end-to-end, and help define how AI systems are built responsibly in regulated domains.
If you enjoy designing clean APIs, scaling Python services, and building systems that actually ship to paying c ustomers, this role is for you.
Compensation: £80,000–£100,000 base + EMI equity
Location: London
About RevaxRevax is building a new class of AI-native software for tax advisory firms.
Today, professional services teams spend countless hours drafting, reviewing, and re-drafting work - constrained by fragmented knowledge, manual workflows, and brittle tooling. We turn a firm’s own expertise into structured, auditable, AI-powered systems that draft, analyse, and explain deliverables with full provenance.
We’re a fast-moving, well-funded startup backed by leading angels, already working with major firms, and building at the intersection of AI agents, knowledge graphs, and real-world enterprise workflows.
What You’ll Do Build & own core backend systems· Build and optimise databases for complex, real-world data
· Design and implement APIs using FastAPI for AI workflows and product features
· Build and scale LangGraph-based multi-agent systems with checkpointing, retries, and observability
· Integrate and scale vector search, knowledge graphs (Neo4j), and associated retrieval pipelines into production services
· Deploy and operate GPU-backed services for AI workloads (e.g. inference, embeddings, or agent pipelines) in production
Scale & operate production infrastructure· Deploy services on GCP (Cloud Run / GKE) or equivalent cloud infrastructure
· Improve performance, reliability, and cost efficiency (latency, throughput, retries, backoffs)
· Implement security-first patterns: auth flows, service-to-service auth, IAM, and access controls
· Help prepare the platform for enterprise customers (SSO, auditability, compliance ready patterns)
Collaborate closely with founders & product· Shape system architecture and API design alongside full-stack and ML engineers
· Translate ambiguous product ideas into robust backend implementations
· Iterate quickly based on real customer usage and feedback
What We’re Looking ForEssential
· 2-3 years building backend systems for real users
· Production experience with FastAPI (or similar modern Python frameworks)
· Solid understanding of API design, PostgreSQL, and distributed systems fundamentals
· Experience deploying and operating services on cloud infrastructure (GCP, AWS, and Azure)
· High ownership mindset: you care deeply about reliability, correctness, and outcomes
Strong Plus
· Experience with LangGraph, RAG pipelines, vector databases, or knowledge graphs
· Familiarity with Neo4j or graph-based data modelling
· Experience working on AI-native or agentic systems
· Comfort working in early-stage startups with ambiguity and speed
· Experience deploying and working with GPU services for AI workloads
Bonus
· Exposure to professional services, finance, tax, or regulated workflows
· You actively use modern AI coding tools (Cursor, Claude Code, etc.)
· Strong instincts for developer experience and internal tooling
Why Revax· Real ownership over core backend infrastructure
· Work directly with founders on high-impact technical decisions
· AI-native product with real customers and immediate feedback
· Modern stack: Python, FastAPI, LangGraph, Postgres, Supabase, Neo4j, GCP
· Meaningful early-stage equity: we hire owners, not passengers
· A lean, low-ego, high drive team that ships fast and cares about their work
Practical details· London based (strong preference for in office collaboration)
· Visa sponsorship available