Member of Technical Staff, Software

Substrate is building the critical infrastructure layer between AI and biology: an AI-native automated lab that produces biological data at scale.

AI for biology has a data problem, not a compute problem. Biological foundation models can predict, but they cannot run experiments, and the high-quality, large-scale data they need does not exist. Substrate generates it, with quality and provenance built in.

We are venture-backed, building our first lab at 20 Triton Street in London, with US expansion to follow. What started as four co-founders is now a rapidly expanding team across science, intelligence, software, operations and partnerships, with people who have come from Automata, Palantir, Owkin, Illumina and Exscientia. We expect to be over 30 people within the year.

We are not a cloud lab and we are not a CRO. We are the infrastructure that turns scientific intent into executed experiments and structured, AI-ready data, and over time into proprietary datasets and our own infrastructural intelligence.

Most software roles like this build a product that lives entirely on a screen. This one runs a physical laboratory. The workflows you orchestrate move real liquid, real cells and real instruments, and the data you capture is the product, not telemetry about it. When something deviates on the floor, your software is what catches it and what records why. Our office sits beside the lab, so you can spend as long as you like watching the automated benches and the transport rails run.

You will also work at the same table as software, hardware and biology, and the three do not always agree. Some engineers find that mix energising; some find it distracting. It is worth knowing in advance which one you are.

You will build the infrastructure software that runs the lab, working across the stack with the founding software engineer and the team.

There are two products. The execution product turns a customer's intent into executed lab work: a translation layer converts an experiment into versioned, runnable workflows, an orchestration layer schedules and runs them across the lab on top of Automata's LINQ, and the output lands as structured, AI-ready data under a shared ontology. The observation product captures metadata everywhere it is generated and maps it into a knowledge graph, so every run carries full provenance.

Where you land depends on you and on what the lab needs next. Any of these could be yours:

Whatever you own, you own it end to end. You will write production code from your first weeks, help set the architecture and the engineering culture, and work at the boundary with the scientists running the assays and the intelligence team who learn from what the lab produces.

You are a generalist who has shipped production systems that other people depend on. You write good code at speed, you have opinions about architecture, and you have learned when to hold them and when to defer. You are happy owning a service end to end, including the parts that are not glamorous: reliability, observability, the on-call pager. You have worked across the stack and can pick up whatever the problem in front of you needs.

You do not need a biology background and we will not test for one; the science is something you will learn enough of by working next to it. What we do want is curiosity about what this infrastructure makes possible, and what it means for the people who will use it.