AI Platform engineer

We're scaling the backend systems powering our AI-driven product suite, including retrieval-augmented generation (RAG) pipelines, real-time inference APIs, and vector search infrastructure. You'll own the services that sit between our LLM providers and our customer-facing products, focusing on reliability, latency, and cost efficiency at scale.

What You'll Do:

• Design and maintain APIs serving LLM-powered features (chat, search, agentic workflows) to 500k+ monthly active users
• Build and optimise RAG pipelines, including chunking strategies, embedding generation, and vector store management
• Architect event-driven microservices using Kafka/SQS for async processing of high-volume inference requests
• Implement observability and cost-tracking for token usage across multiple LLM providers (Anthropic, OpenAI, open-source models via vLLM)
• Own database performance for both relational (Postgres) and vector (pgvector/Pinecone) workloads
• Collaborate with ML engineers on model-serving infrastructure and prompt-caching strategies
• Drive API versioning and backwards compatibility as the product iterates quickly

Core Tech Stack:

• Languages: Python (FastAPI), Go, TypeScript (Node.js)
• Databases: PostgreSQL, Redis, Pinecone/pgvector
• Infra: AWS (ECS, Lambda, SQS/SNS), Docker, Kubernetes, Terraform
• AI/ML tooling: LangChain/LlamaIndex, vLLM, Anthropic & OpenAI APIs, embedding models
• Observability: Datadog, Grafana, OpenTelemetry
• CI/CD: GitHub Actions, ArgoCD

• 4+ years backend development experience, ideally with at least 1 year working with LLM/AI-integrated systems
• Strong grasp of asynchronous architecture and distributed systems fundamentals
• Experience with vector databases or semantic search implementations
• Comfortable working directly with LLM APIs (rate limits, streaming responses, token cost optimisation)
• Solid understanding of API design, security, and scalability best practices

Nice to Have:

• Experience fine-tuning or evaluating open-source models
• Familiarity with agentic frameworks (function calling, tool use, MCP)
• Prior startup/scale-up experience