Head of Machine Learning Engineering

Ironsite AI•San Francisco, CA

6d•$200,000 - $330,000•Onsite

About The Position

Ironsite is hiring a Head of ML Engineering to own the systems and infrastructure that take research models from testing through production. The role will build systems for evaluating models on live jobsites, tools for nightly labeling, data tooling for research, product features powered by machine learning, and the orchestration and inference stack that delivers reports by 5 AM. The position begins as a hands-on technical lead and is expected to grow the team.

Requirements

At least 7 years of software engineering experience, including at least 3 years building production ML systems.
Experience leading a small team as a tech lead or manager while continuing to ship substantial code.
Experience building production experimentation or A/B testing systems.
Experience building internal tools used daily by non-engineers, such as labeling or operations tools.
Hands-on experience with MLOps and GPU inference.
Ability to debug across the stack, tracing customer-report issues to models, data, or pipelines.
Clear, concise communication with researchers, labelers, and customers.

Nice To Haves

Experience with video data at scale.
Experience with labeling tooling, including pre-labeling, active learning, or QA workflows.
Frontend or product experience.
Experience working alongside a research team at an ML startup.
Experience with data from device fleets or other hardware.
Familiarity with construction or other physical-world operations.

Responsibilities

Build production experimentation systems that let researchers A/B test models on live jobsites, compare outputs, and roll out winning models.
Own Tagging Interfaces, including model pre-labeling, prioritization of footage for labeling, and tracking label throughput and quality.
Develop research data tooling for dataset versioning, searching footage and labels, and assembling training sets.
Ship ML-backed features in superintendent reports, investigate poor customer reports across models, labels, and pipelines, and get models working at new jobsites during onboarding.
Own the nightly pipeline orchestrator, model deployment, versioning, rollback, and inference.
Monitor output quality by site and identify drops before customers do.
Manage GPU and cloud costs and accommodate footage and device changes from the hardware team without disrupting the pipeline.
Write code hands-on while hiring and building a team.