Senior Machine Learning Engineer

Senior ML Engineer, Recommendations | Join the team behind a $3.2B smart-home exit, now taking on food waste | Bay Area (Hybrid) | $210K–$240K base

We've partnered with a Bay Area hardware company on one of the most exciting missions in climate tech: keeping food waste out of landfill. The pedigree is serious: the founder previously created one of the defining products of the smart-home era, a company acquired by Google for around $3.2 billion, and has brought the same design obsession to food waste. Their connected appliances turn food scraps into shelf-stable grounds, and the logistics keep it out of landfill and back into use. The market has responded: the consumer product is already generating eight-figure revenue, and with a commercial launch landing in 2027, next year is set for real growth. They are backed at Series C by a heavyweight group of climate and tech investors, including Breakthrough Energy Ventures and Amazon's Climate Pledge Fund.

This is a rare opportunity to own a customer-facing recommendation system end to end: the signals and models that decide what to recommend, the feedback loop that tells you whether it worked, and the improvement cycle that keeps it getting better. You'll sit on the Data team alongside Data Platform, Integrations and Warehouse, and partner closely with product and engineering to make recommendations useful, accurate and genuinely able to change customer behaviour. If you want to own a recommender outright rather than contribute a slice of someone else's, and see your models land in a product with a real purpose, this is the seat.

The Role

As the Recommendations Engineer, you will:

• Build and operate the customer-facing recommendation engine that turns food-waste data into actionable outputs: purchasing suggestions, anomaly explanations and operational nudges, including LLM-based logic where it makes sense.
• Train, evaluate and iterate on models for anomaly detection, pattern recognition and recommendation on production food-waste data.
• Design the features and signals, from IoT sensors to usage data, that feed your models.
• Define and track the metrics that measure whether recommendations actually matter to customers, not just whether the pipeline ran.
• Bring CI/CD and experimentation discipline to model changes: automated testing, staged rollout, A/B testing or holdouts, clear rollback paths, and monitoring that catches degradation in production.

About You

• You have designed, built or operated a recommendation system in production that combines multiple sources into a single customer-facing output, not just contributed data to someone else's model.
• You have trained and evaluated models for anomaly detection, pattern recognition or similar applied ML problems in production.
• You have built recommendation or personalisation logic using LLMs, such as prompt-based scoring, retrieval-augmented generation or agent reasoning, in a live product.
• You are comfortable in production Python and SQL to source and prepare model inputs, and you ship model changes with CI/CD discipline, measuring real impact.
• You have around five or more years in applied ML, recommendation systems or closely related work, and a genuine bias toward action.
• Nice to have: anomaly or fraud detection and forecasting; computer vision or IoT sensor data as a model input; feature stores or ML feature pipelines; and tools such as Hex, Mixpanel or Tableau.
• You are based in the Bay Area and comfortable with a hybrid pattern.

The role pays a competitive base plus equity, with an unusually candid twist: instead of negotiating, you're shown two versions of the package, one weighted to salary and one to equity, and you pick the mix that suits you. The team is moving quickly with this hire. If owning the recommendation system behind a product keeping food waste out of landfill appeals, apply now or send your CV directly to will@inventurerecruitment.com.