About Natural Negative

Natural Negative is a materials science company optimising biomaterials science with AI. We design compostable formulations that match or surpass conventional plastics: production-ready and drop-in compatible with existing manufacturing lines.

We believe everything humans encounter and consume should make them healthier. The biomaterials that could deliver that already exist; the industry hasn't yet made them practical to adopt at scale. We're optimising that: faster, cheaper, less risky.

We're investor-backed, with active Innovate UK grants and partnerships across leading research and innovation organisations. Our lab is at Plus X Innovation Hub, Brighton.

The role

As our ML engineer, you'd own our cloud ML infrastructure, building it out closely with the CTO, who leads the technical direction. You'd work alongside our full-stack engineer to get ML capabilities into the product, and with the materials scientists running the experiments your models train on.

In practice:

Stand up and run our production ML stack on GCP.
Design and automate the pipelines that move experimental data out of the lab and into training.
Bring real engineering discipline to the models: CI/CD, testing, dataset and checkpoint versioning.
Help shape how it all fits together, alongside the CTO.
Contribute to the next generation of our modelling with the founding team.

You'd be early enough to shape how we build, and the systems you stand up will support the platform as it grows. And you'd be working with materials science data, closer to the underlying science than most ML roles ever get.

How we work

We say what we think, kindly. We move thoughtfully. Sprint culture doesn't solve materials science problems. We're here because the problem is interesting and the science is unfinished.

Must have:

Production ML experience: you've built and run ML systems before and can own your delivery, with the CTO setting the wider technical direction.
Strong Python and production ML experience (PyTorch or similar), with hands-on containerised deployment to GCP or comparable platforms.
A track record of building and running ML/data pipelines in small teams, comfortable when not much infrastructure is handed to you.
A systems mindset: you decouple cleanly, think in modules, and can hold the shape of a whole pipeline in your head.
Genuine interest in the underlying problem. You'd be working with materials data; if that doesn't sound interesting, this isn't the role.

Nice to have:

Experience applying ML in scientific, materials, or engineering domains.

How you work:

You take ownership and finish what you start.
You have opinions and you say them, but they're rooted in considered thinking rather than reflex.
You're a good colleague: friendly, empathetic, and able to operate in a flat team.

A competitive salary + meaningful equity.
A direct working relationship with the founders. No layers between you and the people making decisions.
28 days holiday plus bank holidays.
Workplace pension scheme.
Remote working, with lab and studio space at Plus X Innovation Hub, Brighton, if you want it.
A team that doesn't reward presenteeism and doesn't micromanage. Family, rest, and time to think are how good work gets done, not what gets traded to do it.

Natural Negative remote jobs

MLOps/ ML Engineer

Full Stack Engineer