Machine Learning Engineer

Why join us

Meuze is the AI operating system for QSR chains. We forecast what each location will sell and what labor it will need, then we run the work that follows from that forecast: ordering, inventory, waste, and invoices. Today those systems sit in silos. We put them on one platform so a multi location restaurant group can actually run as one business.

We are early, and already moving real money. Eight weeks after launch we hit $9M in contracted annual revenue, with more than 2,000 locations live. Now we are expanding into enterprise groups that run thousands of stores. The contracts are real. The open question is how far we can take it.

You will join a small team with a high bar and real pace. You pick up work that matters instead of waiting for it to be assigned. You work directly with the founders. What you ship goes to people running restaurants through a dinner rush, and they will tell you quickly if it helped.

Engineering at Meuze

Engineering here covers forecasting, agents, product, and the infrastructure underneath. You own what you build from design through production. A lot of this has not been done in QSR before, so there is rarely a playbook to copy. We treat engineering as a craft, and the people who build well here become the people who lead.

What you will do

Time series forecast accuracy is the number this company is judged on. It shapes what every location orders, what it wastes, and what it earns. At enterprise scale, a single point of accuracy is real money. We are hiring a Machine Learning Engineer to own that number: the models, the evaluation, and the MLOps and DevOps that let us improve with confidence. You should read the literature, have a point of view on where forecasting is going, and know the difference between a result that looks good on paper and one that holds up in production.

The role

• Own time series forecast accuracy, and set the standard for what good looks like in this category.
• Advance demand models across seasonality, local effects, and sites with little history, including probabilistic and hierarchical approaches.
• Build evaluation and experimentation so we can tell whether a change actually helped.
• Own MLOps and DevOps for the forecasting platform: training pipelines, versioning, CI, controlled rollout, monitoring, and rollback.
• Explain modelling decisions to stakeholders who will never read the paper behind them.

Bonus points

• A paper or public benchmark result that others in the field took seriously.
• You founded a company or product and took it to real users, revenue, or a team.
• Forecasting experience in retail, logistics, or another domain where demand is noisy and hierarchical.

Apply

We read every application. Attach a resume, and tell us something it cannot.

Apply for this role