[Hiring] Senior Machine Learning Engineer @Numa

Role Description

We’re hiring a Senior Machine Learning Engineer to build and ship ML/AI systems that interact with real customers thousands of times a day. Our voice agents book service appointments, rescue missed calls, and route callers through natural conversations. You’ll work across products and platforms. You’ll ship AI features including prompts, agents, tools, and production ML models, while building the evaluations and tooling that help teams ship with confidence. You’ll also contribute to our ML platform, including model serving, LLM infrastructure, and production observability. At Numa, we believe great ML is about more than building bigger models. It’s about knowing whether a change is good enough to ship. Our evaluation-first approach makes that measurable in CI and production.

What You’ll Do

• Build conversational AI systems for phone and SMS that understand customer needs, take action, and know when to act autonomously.
• Develop tooling such as memory, knowledge graphs, and validated customization that help agents reason and adapt to dealership needs.
• Train, evaluate, and deploy ML models using Ray Serve and Dagster for prediction, classification, ranking, and capacity forecasting, and keep them healthy in production.
• Create offline and online evaluations, simulations, and CI gates that catch regressions and continuously measure quality in production.
• Implement shared evaluation, observability, and LLM tooling that helps product squads ship faster and more reliably.
• Raise the engineering bar through design and code reviews, technical writing, and mentorship.
• Work autonomously when goals are clear and help create clarity when they are not.

Qualifications

• 6+ years of software or machine learning engineering experience, with a track record of shipping ML or LLM powered systems to production.
• Strong Python and solid software engineering fundamentals; you write code others can build on, test, and trust.
• Hands-on experience building AI systems; training and deploying ML models (prediction, classification, ranking) and/or working with LLMs (prompting, tool use, agents, retrieval) with a real intuition for where they break.
• A rigorous, evaluation-first instinct, you reach for measurement to tell a real improvement from a lucky one, and a real regression from noise.
• Comfort operating with meaningful ambiguity in a product environment, and the judgment to know where to invest depth versus speed.
• Ownership and communication; you can carry a piece of work from problem framing through shipped, monitored, and improved.

Nice To Have

• Experience building or operating ML platform and tooling: evaluation frameworks, model serving, feature/prompt registries, or ML observability.
• Familiarity with our stack or its neighbors: LiveKit, Ray Serve, Dagster, Vertex AI, GCP, Kubernetes, Pulumi, or provider APIs (Anthropic, OpenAI, Deepgram, ElevenLabs).
• Comfortable taking a model or agent from prototype to production and owning it there.
• Proficient with real-time or streaming systems: voice, telephony (SIP/WebRTC), or low-latency inference.
• Familiarity working in a startup or high-growth environment with evolving conditions.

What Success Looks Like

• Your squad ships AI features faster and with more confidence because the evals and tooling around them are solid.
• No AI change you own reaches customers without an evaluation standing behind it and regressions are caught before they ship.
• You've turned at least one hard-won squad lesson into platform leverage the rest of the team now relies on.
• Production issues in your area are understood, measured, and closed with statistical confidence that the fix actually worked.
• Other engineers are better at building AI systems because of your reviews, your writing, and how you work.

Benefits

• Base Salary Range: $200,000 - $250,000 CAD.
• Equity Packages.
• Flexible PTO.
• Fully Covered Group Insurance.