Lead ML Engineer — Production AI Agents

Arcade; San Francisco (CA); On-site; USD 230,000 - 260,000; Full time

Job summary

Arcade is hiring a principal machine learning engineer to build the future of agent capabilities. You will own the training pipeline end-to-end, from data intake to model release, and lead models for tool selection, routing, retrieval, and memory.

You will expand use cases to new territory, ship on-prem, and drive the ML stack strategy in a fast-moving enterprise environment. You’ll work with a team of experts to deploy models across enterprise systems, quantify performance with rigorous evals,

Qualifications

• 7+ years of software engineering experience, with 4+ years training and shipping production ML systems.
• Experience training or fine-tuning models that went to production and moved a metric customers cared about.
• Expertise in production agent systems: harnesses, memory, skills, tool use, and sub-agents.
• Know how and when fine-tuning works and how to derive it from raw telemetry data.
• Evaluations you’d defend in a review with data behind decisions.
• Deployment under real constraints like latency budgets and varied infra (vLLM, ONNX, TensorRT, etc.).
• Strong Python for training and ML work, plus TypeScript or Go for production services.
• You pick the stack, write the docs, and defend decisions long-term.
• Comfort with ambiguity in early-stage environment.
• Insatiable desire to ship.

Responsibilities

• Own the training pipeline: end-to-end from data to release.
• Build models for tool selection, routing, retrieval, and agent memory.
• Expand use cases to new territory like search and tool context.
• Measure what matters: offline evals, online production measurement, head-to-head comparisons.
• Ship on-prem: quantize, optimize, and package models for customer VPCs.
• Turn telemetry into training data within enterprise data boundaries.
• Set the direction: choose the ML stack and help hire next in ML.
• Decide what to build using deep knowledge of agent systems.
• Use AI to compound your own output; ship improvements quickly.

Skills

Software engineering

Production ML systems

Agent systems

Fine-tuning models

Model evaluation

On-prem / enterprise deployment

Python

TypeScript or Go

Shipping fast

Ambiguity tolerance

Job description

Arcade is hiring a principal machine learning engineer to build the future of agent capabilities. You will own the training pipeline end-to-end, from data intake to model release, and lead models for tool selection, routing, retrieval, and memory.

You will expand use cases to new territory, ship on-prem, and drive the ML stack strategy in a fast-moving enterprise environment. You’ll work with a team of experts to deploy models across enterprise systems, quantify performance with rigorous evals,