LLM Platform Engineer/Lead (m/f/d)

The Team

The Platform Engineering team owns Real-time bidding (RTB) platform, a Real-time budget system, Real-time event processing, Stream data processing engine and multiple other complex large-scale distributed components and data pipelines. We work closely with data scientists, data analysts, product and business teams. We are responsible for some of the most technically challenging work, for example:

• Handling ~2 Million Req/sec in sub-millisecond latency
• Handing & managing ~Petabytes of data
• Managing distributed systems deployed across multiple data centers
• Optimizing JVM and Linux kernel for optimal performance
• Managing our own data center spread across the globe consisting of thousand of powerful servers
• Working on some of the most challenging problems of Ad-Tech

About the Role

We are looking for an LLM Platform Engineer / Lead who can evangelise AI across the company and build the foundational tooling, frameworks, and workflows that allow teams to ship AI-powered features quickly and safely. This role is ideal for an experienced engineer who understands modern AI capabilities and focuses on developer velocity, reuse, and production readiness rather than training complex models from scratch.

Responsibilities

• Design and build internal AI frameworks, SDKs, and shared libraries
• Enable teams to integrate AI features with minimal friction
• Set up standardized patterns for using LLMs, embeddings, agents, and workflows
• Build reusable components for prompt management, evaluation, observability, and safety
• Define best practices for AI usage, cost control, and reliability
• Evangelise AI internally through documentation, examples, and hands-on guidance
• Rapidly prototype AI-powered features and turn them into reusable building blocks
• Own AI tooling from experimentation to production

Requirements

• Hands-on experience with LLM APIs (OpenAI, Anthropic, etc.)
• Experience with RAG pipelines, embeddings, vector databases
• Familiarity with prompt engineering, prompt versioning, and evaluation
• Experience with AI orchestration frameworks
• Good understanding of AI observability, cost monitoring, and failure modes
• You’ve shipped at least one LLM-powered feature to production and iterated based on telemetry or user feedback
• Comfortable with embeddings, fine-tuning, vector search, tokenisation, and evaluation methodologies
• Familiar with data