AI / ML Engineer

WareBee Inc.; United Kingdom; Hybrid; GBP 70,000 - 120,000; Full time

Benefits offered by this job

Small senior team

Hybrid in Cambridge or Tel Aviv

Fully remote option

Job summary

WareBee is seeking an AI/ML Engineer to own the full lifecycle—from framing the problem with the team to training, evaluating, and shipping models into production where operators rely on them daily.

You will design data pipelines turning real-world warehouse inputs into reliable signals, ensure observability with clear rollback options, and pair with engineers to keep the codebase readable. Hybrid work options in Cambridge, UK or Tel Aviv, Israel, or fully remote.

Qualifications

• Frame problems with the team and design data pipelines for training and inference signals.
• Train, fine-tune, and evaluate models in production context.
• Ship models with observability and rollback mechanisms.
• Experience taking a notebook model to production.

Responsibilities

• Frame the learning problem with the product team.
• Design data pipelines turning operational inputs into training and inference signals.
• Train, evaluate, and ship models with production-grade observability.

Skills

Deep learning

Probabilistic methods

Production ML

Python

Readable code

Job description

At WareBee, learning systems are not a bolt-on — they are how the digital twin turns raw warehouse data into decisions managers can act on. As an AI / ML Engineer you will own the full lifecycle: framing the problem with the team, shaping the data, training and evaluating models, and shipping them into production where real operators will lean on them every day.

WareBee runs on two engines: Physical AI — a living, spatial model of the warehouse floor, its racks, aisles, and movement — and Process AI, which learns how work actually flows through it. The learning systems you build sit at the seam between the two.

What you will do

• Work with the product team to frame what a learning system can usefully solve in a warehouse and what it cannot
• Design data pipelines that turn messy real-world operational inputs into reliable training and inference signals
• Train, fine-tune, and evaluate models against the actual problem — not against a leaderboard
• Ship models into production behind clear contracts, with observability and a way to roll back when reality disagrees with the validation set
• Pair with engineers to keep the surrounding application code legible and maintainable

What we are looking for

• Strong fundamentals in modern machine learning — deep learning, probabilistic methods, evaluation discipline
• Experience taking a model from notebook to production with all the unglamorous parts in between
• Comfort writing software that other people will read, debug, and modify
• A bias toward measuring whether a system actually helps users, not whether the metric goes up
• Curiosity about logistics and operations, not just the technique
• Hands-on warehouse, logistics, or supply-chain experience — you’ve seen how a real floor runs (a strong plus)

What we offer

• Product ownership from problem framing through deployment
• A small senior team — no committees, no review boards, no theatre
• Hybrid in Cambridge, UK or Tel Aviv, Israel — or fully remote