ML Engineer
Through Indeed and our own Indeed Flex agency, we see labor-market activity at a scale no single agency can: workers looking for work, employers creating demand, shifts being worked.
Day to Day
As a Staff Machine Learning Engineer in our Staffing Technology Marketplace group, you'll own the source data, the models and the business impact behind Indeed's AI-first staffing platform, which agencies, employers and the MSPs that coordinate them use every day to take orders, book workers and manage shifts. Through Indeed and our own Indeed Flex agency, we see labor-market activity at a scale no single agency can: workers looking for work, employers creating demand, shifts being worked. You'll set the technical direction for turning that into models that predict whether a booked worker will turn up, whether and how fully an order will fill, and which orders are at risk early enough for someone to act. You'll then turn those predictions into decisions customers act on and increasingly trust the platform to make for them: how many backups to book, which agencies an MSP should release an order to and in what sequence within contracted volume splits, and what pay rate will achieve the fill and quality a client asked for without overspending.
On a typical day you'll work with customers and our business, product and engineering teams to define, build and evaluate production models per market in the US and UK. You'll keep them healthy and catch a model degrading for a single customer before they notice. You'll design ranking and reliability scoring that balances today's fill against the long-term health of the worker pool, so the same few workers aren't overused while everyone else drifts away. You'll make models work from day one for a new agency with no history, and hold up on the thin data most agencies have. Every recommendation needs to be explainable to the person acting on it and configurable to each agency's risk appetite. You'll use what the models show to shape the roadmap with product, engineering and design, and work with legal and engineering partners on foundational questions, such as what can be learned across agencies and what must stay separate.
Responsibilities
Posted November 7, 2026