ML Senior (AI / LLM, Gaming Industry)

About: The team is building an AI platform in the gaming industry, working with LLMs, RAG, and multi-agent systems in production.

Responsibilities:

• Design LLM architecture: RAG, multi-agent, tool-calling, memory
• Scale/optimize vector & semantic search (pgvector, hybrid search, reranking)
• Build pipelines for GDD processing and validation
• Develop prompt engineering, evaluation, and guardrails
• Integrate local and cloud LLMs (Ollama / OpenAI-compatible APIs)
• Design fault-tolerant AI systems for production workloads
• Implement monitoring for quality, latency, and cost
• Architect multi-agent workflows and manage agent coordination
• Prepare the platform for TypeScript code auto-generation
• Code review, mentoring, and team technical growth

Must-have:

• 3+ years in ML / Applied AI / LLM projects
• Strong Python
• Hands-on LLM, RAG, embeddings, vector search
• Production AI/ML system design & maintenance
• Multi-agent architectures and LLM orchestration
• PostgreSQL + pgvector (or similar)
• Async and event-driven pipelines
• Docker (production)

• Architectural mindset and ownership

   

Nice-to-have:

• HuggingFace / vLLM / OpenAI / Ollama
• Hybrid search, rerankers, long-context strategies
• RabbitMQ / Redis
• CI/CD and MLOps practices
• TypeScript (basic understanding)
• Experience in game dev or content generation
• Confluence / Jira API integrations