Nyxium
Software Engineer
Get notified of more jobs like this · No spam, ever
How your CV stacks up
Upload your CV to see how well it fits this job role
?%
Nyxium
Nyxium turns fragmented, consultant-led infrastructure diligence into a single decision-ready system. When someone's trying to site a data centre, BESS project, or energy asset, the real question is deceptively hard: can this exact plot of land actually get power, planning permission, and grid connection, in a timeframe that makes the project viable. Today that question gets answered by consultants stitching together PDFs, spreadsheets, GIS layers, and phone calls. We're building the workspace that answers it directly, with full source traceability behind every conclusion.
We're early-stage, venture-backed, with a live pipeline and a major launch ahead.
The Problem You'd Be Working On
Grid, land, planning, and environmental data are messy, inconsistent, and often contradictory across sources, and none of them were built to be machine-readable in the first place. Our agents have to pull from these sources, reason over conflicting or incomplete evidence, and produce a confident, explainable output that a developer or investor can actually act on and trace back to its source.
Reasons to use Rodeo
Start with a chat, not a search bar
Grad scheme, placement, apprenticeship? Not sure what you want yet — that's fine. Your agent talks it through with you and turns "I have no idea" into a shortlist.
It searches the market for you
Every day your agent scans the market matching roles against what actually matters to you, not just keywords on a CV.
Only hits
No noise. No "maybe this fits." Just roles with a clear explanation of why they're right — and where to focus when applying.
That's a genuinely hard agent design problem: how do you get an LLM-based system to know when it doesn't have enough evidence, surface its uncertainty instead of hallucinating a clean answer, and keep a full provenance trail from raw data to final recommendation. You'd be working directly on that, not bolting AI onto an existing workflow, but building the core reasoning and evidence layer of the product itself.
What You'll Actually Do
- Ship in TypeScript day to day, reach for Python when a problem calls for it, and build on Google Cloud infrastructure.
- Design and tune agents that pull structured and unstructured data (planning documents, grid capacity data, environmental records) and reason over it reliably.
- Build the evidence and confidence layer that lets every output trace back to its source, this is core to the product, not a nice-to-have.
- Work on the boundary between clean software engineering and genuinely uncertain, adversarial real-world data.
- Have direct input into technical architecture decisions with our CTO, this early, that's a real say, not a token one.
Get help with your application
Your very own career expert that helps elevate your application to the next level.
What We're Looking For
- Strong fundamentals: you write clean, secure, maintainable code, not just prototypes.
- Real experience with LLM agents in production, including where they fail and how you've handled that (hallucination, inconsistent output, unreliable tool calls).
- TypeScript day to day, comfortable in cloud-native environments, ideally GCP.
- Someone who's shipped things end to end and is comfortable getting on with it without a fully specced backlog.
- Based in or willing to relocate to London.
Genuinely Interesting If You've Worked With
- Probabilistic graphical models, Bayesian networks, or decision theory, we're dealing with uncertainty and conflicting evidence constantly, this isn't a box-tick, it's close to the actual problem.
- Knowledge graphs or ontologies, useful for how we structure fragmented data.
- Constraint-solving or optimisation.
- Explainable AI, provenance matters as much as the output itself.
“It took my CV and asked me questions relevant to understanding what kind of jobs to suggest for me. Suggestions were almost perfect. Jobs were exactly what I’ve been looking for.”
Jessica, London
Skills
Location