ML/AI Engineer — Multi-Agent Analytics on Databricks

Harnham; United States; On-site; USD 120,000 - 180,000; Full time

Job summary

Harnham seeks a highly skilled AI/ML Platform Engineer to join a large enterprise building a multi-agent analytics platform that enables users to query data through natural language and receive trustworthy, explainable answers at scale.

You will design and operate a system of cooperating LLM-powered agents that query data sources, run analyses, generate visualizations, and deliver production-grade insights. Collaboration with engineering and product teams is essential to success.

Qualifications

• Deep expertise with Databricks platform and production AI systems.
• Experience designing and operating multi-agent AI workflows.

Responsibilities

• Design and implement multi-agent workflows within a Databricks-based analytics platform, including planning, routing, tool calling, memory, and handoff between specialized agents.
• Build and maintain agent tools including data source connectors, retrieval pipelines, SQL generation and validation, visualizations, and result verification.
• Develop evaluation frameworks, testing suites, and regression tracking to improve quality and reliability.
• Improve answer accuracy and groundedness through guardrails, validation mechanisms, and explainability.
• Optimize for latency, cost, throughput, and production resilience.
• Partner with engineering, analytics, and product teams to deliver trusted AI-powered analytics experiences.

Skills

Databricks platform

Python engineering

LLM / agentic apps

OpenAI Agents SDK

Multi-agent architectures

Tool calling

RAG

CI/CD pipelines

Cloud-native deployments

Tools

Databricks Apps

Databricks SDK

Databricks Asset Bundles

Databricks APIs

LangGraph

OpenAI Agents SDK

Vector Search

MLflow

CI/CD

Cloud-native deployments

Job description

Harnham seeks a highly skilled AI/ML Platform Engineer to join a large enterprise building a multi-agent analytics platform that enables users to query data through natural language and receive trustworthy, explainable answers at scale.

You will design and operate a system of cooperating LLM-powered agents that query data sources, run analyses, generate visualizations, and deliver production-grade insights. Collaboration with engineering and product teams is essential to success.