Software Engineer (AI Platform) - London

Revax; Greater London; On-site; GBP 65,000 - 110,000; Full time

Job summary

Revax is building an enterprise-grade tax research platform in London, inviting engineers to own features end-to-end across the stack. You will collaborate with the founders to design clean APIs, scale services, and ship to paying firm customers.

The role emphasizes reliability, producing robust AI-driven workflows, and contributing to regulated-domain projects from day one.

Qualifications

• 2–3 years building production software.
• Experience with LangGraph, RAG pipelines, vector databases, or knowledge graphs.
• Production experience with FastAPI or a similar modern Python framework.
• Experience scaling and operating services on GCP, AWS, or Azure.
• You care about reliability, correctness, and outcomes, not just lines of code.
• Heavy use of modern AI coding tools (Cursor, Claude Code, Codex).
• Experience with multi-tenant deployments for enterprise customers, VPC networking, Kubernetes, or shipping into procurement-heavy regulated buyers.
• Exposure to Neo4j or graph-based data modelling.
• Working knowledge of LlamaParse/LlamaIndex.
• Experience optimising low-latency applications.
• Time spent on AI-native or agentic systems in production.

Responsibilities

• Own features end-to-end across the stack for enterprise customers.
• Design and implement APIs in FastAPI for AI workflows and product features.
• Build and scale LangGraph-based multi-agent systems with checkpointing, retries, and observability.
• Integrate vector search, knowledge graphs, and retrieval pipelines into production services.
• Deploy services on GCP (Cloud Run / GKE) and optimise performance, reliability, and cost.
• Implement auth, IAM, audit, and access-control patterns expected by enterprise buyers.
• Assist with deploying SSO, tenant isolation, auditability, and security reviews for major firms.
• Collaborate with customers, other engineers and founders to shipping best-in-class software.

Skills

Production software experience

Ownership mindset

AI coding tools

Reliability & correctness

Tools

FastAPI

Python

GCP

AWS

Azure

Kubernetes

Neo4j

LlamaIndex

LlamaParse

LangGraph

Vector databases

Job description

The role

We're building out our engineering team to allow us to onboard more firms, and to help us build the leading tax research platform faster. You will work directly with the founders, owning features end-to-end across the stack, and helping to define how AI systems are built responsibly in regulated domains. Customer calls, prototyping, production hardening, and the work that lets us sell into leading firms - all of it is in scope. If you enjoy designing clean APIs, scaling services, and shipping products to paying enterprise customers, this role is for you.

What you'll do

• Model and optimise PostgreSQL / Supabase & graph db / Neo4j schemas for complex data
• Design and implement APIs in FastAPI for AI workflows and product features
• Build and scale LangGraph-based multi-agent systems with checkpointing, retries, and observability
• Integrate vector search, knowledge graphs, and retrieval pipelines into production services
• Deploy services on GCP (Cloud Run / GKE) and improve performance, reliability, and cost
• Implement the auth, IAM, audit, and access-control patterns enterprise customers expect
• Help deploy SSO, tenant isolation, auditability, and the security reviews that come with selling to major firms
• Work with customers, other engineers, and the founders to create best in class software

Who we're looking for

• 2–3 years building production software
• Experience with LangGraph, RAG pipelines, vector databases, or knowledge graphs
• Production experience with FastAPI or a similar modern Python framework
• Experience scaling and operating services on GCP, AWS, or Azure
• A high-ownership mindset - you care about reliability, correctness, and outcomes, not lines of code
• Heavy use of modern AI coding tools (Cursor, Claude Code, Codex)
• Experience with any of; multi-tenant deployments for enterprise customers, VPC networking, Kubernetes, or shipping into procurement-heavy regulated buyers.
• Exposure to Neo4j or graph-based data modelling
• Working knowledge of LlamaParse/LlamaIndex
• Experience optimising low-latency applications
• Time spent on AI-native or agentic systems in production

Nice to have

• Exposure to professional services, finance, tax, or other regulated domains
• Strong instincts for developer experience and internal tooling

The package

• London-based, with a strong preference for in-office
• Modern stack, and driven team that you can get have an impact in from day 1.