Platform engineer AI platforms

Playamp is MTG’s Midcore District, bringing together six gaming studios that create games played by tens of millions of people across mobile and PC. It provides a shared ecosystem spanning marketing, data analytics, technology, player services, publishing, D2C distribution, and infrastructure, enabling each studio to focus on building games while benefiting from shared expertise and capabilities.

• Design, build, and operate Playamp’s internal AI platform, including the model gateway, agent orchestration, RAG pipelines, vector stores, and MCP servers connecting LLMs to internal systems
• Productionize AI infrastructure on GCP using Vertex AI, GKE, managed and self-hosted inference, Terraform, and Git Ops
• Bring AI into Dev Ops and automation workflows
• Own the production agent lifecycle, including registry, versioning, observability, tracing, evaluations, cost tracking, and regression gates
• Share standard senior Dev Ops responsibilities, including production ownership, on-call, networking, security hardening, and incident response for the AI platform’s core infrastructure
• Develop guardrails that help security teams track and monitor AI usage across the company

• 5–7 Years of production infrastructure, Dev Ops, or platform engineering experience, including 2+ years dedicated to AI infrastructure involving real systems rather than POCs
• Practical experience with MCP and RAG, including building or integrating MCP servers and exposing internal systems to LLMs
• Experience designing and shipping production agentic systems with multi-step, tool-using agents, guardrails, retries, and evaluation
• Deep cloud experience, preferably GCP, and solid Kubernetes, networking, Infrastructure as Code, Terraform, CI/CD, and Git Ops fundamentals
• Experience building or owning LLM/agent evaluation harnesses, including golden datasets, offline and online evaluations, CI regression gates, and production A/B testing of prompts and agents
• Strong Python skills and hands-on experience with modern LLM serving stacks and at least one industry-standard agent framework, such as Google ADK or Lang Chain
• Experience managing production AI spend, including prompt and semantic caching, model selection trade-offs, batch versus real-time routing, and per-team budgets and showback
• Будет плюсом:
  Cost engineering for AI

• Officially registered full-time employment
• Paid annual leave according to local regulations
• Medical support and paid leave
• Individual development plan and regular feedback
• Professional seminars, workshops, courses, and internal training programs
• Reimbursement of gym membership fees
• Benefits may vary by country