Senior AI Engineer at HFM

on-site · full time · Visa sponsorship

Apply for this role at HFM

Responsibilities:
- Architect the 'Marketing OS' and lead design of autonomous, multi-step agents (LangGraph/LangChain) to automate marketing workflows such as campaign planning and asset auditing.
- Design and build high-performance, asynchronous backends (FastAPI) and integrations to support heavy marketing automation and tool chains.
- Define hybrid model strategies: deploy proprietary APIs (OpenAI/Anthropic) for high-reasoning tasks and open-source models (Llama/Mistral) for cost-efficient processing.
- Implement RAG workflows and long-term memory using PostgreSQL with pgvector and Redis for fast semantic retrieval.
- Implement observability, tracing, and cost controls (Langfuse) and enforce model governance and alignment with brand/compliance guidelines.
- Optimize model serving performance (PyTorch, vLLM) and ensure scalable deployment via Docker and orchestration.
- Collaborate with product, engineering and compliance teams to deliver reliable, auditable AI systems.

Requirements:
- Advanced Python and FastAPI experience; strong engineering background building production APIs and async systems.
- Hands-on experience with LangGraph/LangChain, RAG patterns, PostgreSQL + pgvector, and Redis.
- Experience with PyTorch, transformer architectures and model serving tools (vLLM or similar).
- Familiarity with observability/tracing tooling (Langfuse or equivalent) and practices for model governance and cost monitoring.
- Practical knowledge of deploying hybrid LLM stacks (OpenAI/Anthropic and open-source models) and evaluating trade-offs.
- Strong system design skills, experience building autonomous agents, and ability to work cross-functionally in a regulated environment.