Senior AI Platform Engineer

Our client is a fast-growing European fintech company in the business spend management space, offering corporate cards and related financial products to SME and mid-sized business customers across the EU and UK. It operates in a regulated environment with GDPR compliance obligations.

Engineering is organized into cross-functional squads, with platform and enabling teams providing shared tooling horizontally. The engineering practice is AI-native: Claude Code and GitHub Copilot are used daily, supported by a growing library of shared, versioned Skills embedded in key repositories. This supports an end-to-end flow from ticket to implementation, testing, and pull request across several codebases. AI tool and connector rollout follows an approved-list and pilot process.

The environment is AWS-first overall. Data-platform and analytics workloads also use GCP, with BigQuery as the primary data warehouse. Day-to-day coordination uses Slack, Linear for ticket tracking, GitHub for code and pull-request review, and Notion as the knowledge base.

The client is building its AI platform and agentic capabilities. It has launched an MCP surface in closed beta, enabling external AI assistants to connect and perform structured workflows behind guardrails. Agent reliability is treated as a system property, including tool contracts, approval and consent gates for write actions, and evaluation harnesses.

Key Responsibilities

• Design, build, and operate shared GenAI infrastructure, including LLM routing, vector search and RAG infrastructure, an MCP gateway, and AI observability and evaluation tooling.
• Apply distributed-systems fundamentals, including asynchronous workflows, idempotency, and failure design.
• Work hands-on with LLM APIs in production.
• Apply an AI-security mindset, including prompt-injection defenses, PII handling in logs, and credential handling.
• Apply engineering rigor to the Context Development Lifecycle (CDLC): generate, evaluate, distribute, and observe the context that powers AI agents.
• Increase the output and quality of the squad and share patterns and practices beyond the immediate team to support wider adoption.
• Embed directly into a client squad as a hands-on individual contributor, rather than in a coaching or managerial role.

Must-Have Skills

• Primary proficiency in Kotlin or Python, with comfort contributing to both.
• Experience designing, building, and operating shared GenAI infrastructure: LLM routing; vector search and RAG infrastructure; MCP gateway; AI observability and evaluation tooling.
• Strong distributed-systems fundamentals: asynchronous workflows, idempotency, and failure design.
• Hands-on production experience with LLM APIs.
• AI-security mindset: prompt injection, PII in logs, and credential handling.
• Agent and harness engineering fluency, with concrete evidence of building and operating this type of tooling, not only using it.
• Comfort using Claude Code and GitHub Copilot daily.

Nice-to-Have Skills

• Cloud depth in AWS and/or GCP.
• Grafana experience.
• Experience with named LLM API vendors.

The RFP describes these capabilities generically rather than naming specific tools for this role; the nice-to-have items are plausible but not confirmed.