Software Engineer, Gen AI & Web Apps
•Benchsci· 19h
💻 Engineering⏰ Full-time📈 Mid Level💰 $100k-125k🇨🇦 Toronto, Canada
19h ago
Vow·7 days ago
Vow is a cultured meat company on a mission to feed billions and create food that’s good for your tastebuds, your health and your planet.
We're hiring a software engineer to build a software factory. The best manufacturing companies learned decades ago that you don't just build products, you build production systems. We think the same is true for software now.
Your job isn't to write every line of code. It's to design the codebase and guardrails, embed domain knowledge and feedback loops that let us reliably produce software for bioreactor control, process orchestration, and factory analytics. You'll encode our approach into reusable patterns and build validation systems that reject bad output before it reaches hardware or creates bad data.
This is full-stack in the truest sense. On any given week, you might be debugging a bioreactor control sequence on the factory floor, designing database schemas for process data, generating operator interfaces, or optimising real-time state estimation algorithms. The scope is unusually broad because our team is deliberately small and our ambition is not. You won't be handed a perfectly scoped backlog, you'll help decide what's worth building next, and increasingly, how we build.
You'll report to our Head of Software and work directly with process engineers, scientists, and operators. Your code will run on physical hardware that produces real food.
AI-driven development systems: Skills, commands, subagents, MCP servers, custom harnesses - you'll build the scaffolding that lets us (and non-SW engineers) reliably generate factory software and use our data.
Control systems & orchestration: Python-based process control for bioreactors with safety guarantees. Hardware I/O via Modbus and OPC-UA. Coordinating multi-day production runs across dozens of devices.
Operator interfaces: React frontends for machine control, process visualisation, and alarm management.
Data & Infrastructure: PostgreSQL, MongoDB, Redis, Kafka, AWS. Design schemas, optimise queries, surface process insights.
We ship daily. Small PRs, fast iteration, continuous deployment. The factory runs 24/7, and our software evolves with it.
We're genuinely curious how far AI-driven development can go. Claude Code, Codex, custom agents, MCP - we use them daily and want to push further. If you have ideas about how to 10x your own output, you'll have the freedom to try them here.
This is internal software with fast feedback loops, which means unusual room to experiment.
You'll spend meaningful time on the factory floor, understanding how operators interact with your systems, how devices behave under real conditions, and how process constraints shape requirements.
Collaboration is a constant. You'll work with bioprocess engineers and scientists, operators, data scientists and hardware engineers.
People who do well here enjoy being close to the real world, prefer speed over ceremony, and get energy from clarifying and shaping ambiguity into actionable plans.
They're strong software engineers first - clean abstractions, thoughtful APIs, pragmatic tests - but don't want to be siloed. They like understanding how hardware, operators, and business constraints shape good systems.
You communicate proactively. Working with scientists, operators, and engineers from different disciplines means translating between contexts constantly. And you're drawn to the mission: building software for a cultivated meat factory is harder than building another SaaS product. The constraint is that it has to work in the physical world.
If you want a well-defined lane and a predictable backlog, this isn't the role. If you want to become a better engineer - not just a better software developer - this is for you.
Ownership without gatekeeping. No architectural review board, no months-long approval process. You'll make significant technical decisions, own the outcomes, and iterate based on real-world results. The ceiling is whatever you're ready to carry.
Hardware in the loop. Your code runs on physical devices 50 meters from your desk. Write a control sequence on Monday, watch it run a production batch on Tuesday.
The problems are unsolved. How do you roll back a deployment mid-operation? How do you scale state estimation across dozens of bioreactors? We don't have all the answers, you'll help us find them.
Strong software engineering fundamentals - clean code, sensible APIs, evolvable data models
You've built and operated production systems
Zero-to-one experience: shipped something where little existed before, made tradeoffs, left it maintainable
Strong product instincts - you talk to users, distill problems, iterate
Heavy use of AI coding tools (Claude Code, Copilot, agents) to ship faster
Debugging control logic on the factory floor, then refining the system that generated it
Translating process engineering requirements into validation rules
Reviewing outputs from systems you built, not just code you wrote
Process control concepts (PID, state machines, batch processes)
Software that touches physical systems - or strong curiosity about it
Based at our facility in Sydney, Australia - converted warehouse with lab, factory floor, culinary center.
You'll work on-site at least 4 days per week, the factory is in Alexandria, and being physically present matters for debugging hardware integration, working with operators, and understanding the manufacturing context.
We don't track hours. We track outcomes.
Competitive salary + equity - everyone at Vow is an owner
16 weeks paid parental leave (8 weeks caregiver + 8 weeks birth giver)
Unlimited personal leave
Relocation support for interstate or international hires
