[Remote] ML/AI Engineer
Note: The job is a remote job and is open to candidates in USA. Harnham is a leading recruitment specialist in Data and AI, currently partnering with a major Media & Streaming company. They are seeking an ML/AI Engineer to design, build, and operate intelligent agents for a multi-agent analytics platform that provides reliable and explainable data analysis.
Responsibilities
- Design and implement multi-agent workflows, including planner/executor patterns, tool-calling, routing, memory, and handoff between specialized agents
- Build and maintain agent tools such as SQL generation and validation, data-source connectors, retrieval pipelines, chart/report generation, and result verification
- Develop evaluation frameworks including offline test suites, golden datasets, LLM-as-judge graders, and regression tracking
- Optimize system performance for latency, token cost, throughput, and graceful degradation under failure
- Implement tracing, logging, and observability to ensure agent runs are debuggable and auditable
- Collaborate with data engineers, analysts, and product teams to define and encode analytics correctness into the platform
- Evaluate emerging agent and LLM technologies and make pragmatic build-versus-adopt decisions
Skills
- 5+ years of software engineering experience, including at least 2+ years building ML/AI or LLM-powered applications in production
- Strong Python skills and experience with modern LLM orchestration frameworks such as LangGraph or OpenAI Agent SDK
- Strong demonstrable experience with Databricks Asset Bundles & GenAI tools
- Practical experience with tool/function calling, retrieval-augmented generation (RAG), prompt design, and agentic architectures
- Hands-on experience with agent evaluation frameworks and LLM observability tooling
- Experience with vector databases and retrieval optimization
- Experience optimizing LLM workloads for cost and latency
Benefits
- 401k
- Medical
- Vision
Company Overview