AI Engineer

OTAKOYI is looking for a smart and eager Senior AI Engineer(part-time) to join our team. We like challenges and self-development. If you like it too, don’t hesitate to join us!

What You’ll Do
AI Development & Implementation

• Design and implement multi-agent systems (supervisor-worker, collaborative topologies) with production-grade failure recovery
• Build advanced RAG pipelines: hybrid search, reranking, GraphRAG, Agentic RAG — with justified architectural trade-offs
• Develop scalable backend AI services using Python, FastAPI, Pydantic, and Celery
• Integrate vector databases (pgvector, Pinecone, Weaviate) into production AI systems
• Implement MCP servers and integrations (Streamable HTTP transport, OAuth 2.1)

Eval Design & Production Quality

• Build eval infrastructure as part of CI/CD: golden datasets, regression detection, online monitoring
• Design evaluation metrics per use-case (RAGAS for RAG, custom for agents) and maintain dataset freshness
• Implement production observability via LangSmith / Helicone / Langfuse: traces, spans, A/B model testing, anomaly detection

Cost & Architecture

• Design multi-tier model routing strategies and prompt caching to control LLM inference cost
• Contribute to architecture decisions with documented trade-offs (RAG vs fine-tuning, model selection, ADR)
• Apply Spec Driven Development: formalize AI tasks with eval criteria before implementation

Delivery

• Work within MAE delivery model using Claude Code as primary agentic coding environment
• Deploy and manage containerized AI services via Docker and Kubernetes on AWS / Azure

Required Skills

• Hands-on experience building multi-agent systems with LangGraph; knows failure modes in production (infinite loops, state desync, tool call storms)
• Advanced RAG implementation: chunking strategies, hybrid search, reranking, evaluation metrics (MRR, NDCG, recall@k)
• Production eval pipeline experience: can walk through an eval they designed — metric, dataset, what it caught
• LLM APIs: OpenAI, Azure OpenAI, Anthropic Claude, Gemini — can justify model selection with data
• Prompt engineering: system prompt design, few-shot calibration, prompt caching, injection defense
• LLM cost optimization: token economics, model routing, TCO analysis
• MCP integration: knows spec, Streamable HTTP, can build a custom MCP server
• Safety: OWASP LLM Top 10, prompt injection defense (direct + indirect), guardrails architecture
• Production Python: async, FastAPI with DI + error handling, Pydantic, Celery
• LLM observability: LangSmith / Helicone / Langfuse — traces, alerting, regression detection
• Docker + CI/CD with automated eval runs on PR

Nice to Have

• n8n for AI workflow automation
• GCP / Vertex AI experience
• DSPy for systematic prompt optimization
• MLflow or DVC for experiment tracking
• Experience with BMAD methodology for large-scope projects

Відгукнутись на вакансію