Senior Software Engineer (Backend / AI Systems)

The problem we solve
The security industry has spent years making detection faster - better scanner, smarter alerts, more automation and now AI can find and fix vulnerabilities in minutes.

Yet the same classes of weakness continue to appear in audits, pentests, and incidents.

The problem is not speed, the problem is that nobody is looking at the architecture.

Software systems are not understood, they are assumed. Engineers inherit codebases, services accumulate, trust boundaries emerge across teams, clouds, and CI/CD pipelines. Over time, nobody has a complete picture of how the system actually works.

Veriom changes that - we build a unified model across code, cloud, CI/CD, and production environments, helping organisations understand how their systems are actually structured and where architectural weaknesses create risk.

If that is the problem you want to work on, this role is for you.

Who we are
Veriom is building technology that has never existed before.

We are a small team of engineers, security specialists, and systems thinkers building a new way to understand software systems. Our goal is simple: make software architecture visible, measurable, and understandable.

We are venture-backed and we were selected and sponsored by the UK Government to represent British cybersecurity innovation at RSA Conference in San Francisco.

Most importantly, we are still early.

The architecture is still evolving. The hardest technical decisions are still ahead of us. The people joining now will help define the company, product, and technical foundations for years to come.

Who we are looking for
We are looking for a senior engineer who loves owning problems.

Someone who has built and operated real systems under real conditions, has opinions about architecture they can defend, and is already thinking about what the codebase looks like in two years, not just next sprint.

You will be a key technical voice at Veriom.

Day to day, your work will be primarily backend-focused: LLM pipelines, agentic workflows, distributed processing, graph construction, API design, and platform architecture.

If you have experience in offensive security, security tooling, observability infrastructure, or systems engineering, we would especially like to hear from you.

What you will work on
The backend systems powering Veriom's core analysis engine.

• Knowledge graph construction across code, cloud, CI/CD, and infrastructure.
• Agentic RAG pipelines that reason across complex technical environments.
• Distributed job processing infrastructure supporting long-running analyses.
• APIs consumed by customers, partners, and future platform integrations.
• Architecture decisions that will shape the platform for years.
• Occasional frontend contributions where they help move the product forward.

This is not a role where you inherit a finished system and maintain it. The design is still live.

Must have

• Strong Python fundamentals, including asyncio, FastAPI, and clean API design.
• Production AWS experience.
• Strong PostgreSQL and Redis knowledge.
• Experience owning CI/CD pipelines, ideally GitHub Actions.
• Hands-on experience building LLM-powered systems beyond simple API integrations.
• Experience designing RAG or agentic workflows.
• Experience operating distributed systems and background job infrastructure.
• Strong engineering judgement and the ability to make pragmatic architectural trade-offs.
• Vector databases and similarity search systems.

Nice to have

• LangGraph, LangChain, or equivalent orchestration frameworks.
• Experience evaluating and monitoring LLM performance in production.
• Cost optimisation experience for large-scale LLM workloads.
• Kubernetes and container orchestration experience.
• Experience with observability or developer infrastructure platforms
• Offensive security, application security, or security tooling experience.

What we offer

• Equity: 0.5%–1.0%, depending on experience and impact.
• Salary: £45,000–£55,000.
• Flexible working. We measure outcomes, not hours.
• 25 days annual leave plus public holidays.
• Your birthday off.
• Private health insurance.
• A small team where your decisions directly influence the product and company direction.
• The opportunity to work on genuinely difficult technical problems at the intersection of software architecture, security, AI, and systems engineering.

Every major technical decision is still being made.

If that excites you, we'd love to hear from you.

Help build the thing the rest of the industry will eventually have to catch up to.

Application Questions

We review every application personally.

Please include brief answers (a few paragraphs each, not essays) alongside your CV.

1. Describe a production LLM system you built or owned. What broke in production that you did not anticipate at design time, and how did you respond?

2. A background job makes multiple LLM calls and fails halfway through. Walk us through:

• what state you persist,
• what you would re-run,
• what you would never re-run,
• and how you handle the fact that LLM outputs are non-deterministic.

1. You inherit 15 external API integrations, each implemented differently with no shared interface. What is the first architectural change you make, and why?

2. Tell us about a technical decision you strongly disagreed with on a project. What was your position, what happened, and what did you learn?

3. Optional but valued: have you worked in offensive security, red teaming, security tooling, developer tooling, platform engineering, or observability infrastructure? If so, tell us briefly about the most interesting problem you worked on.

There are no perfect answers to these questions. We are interested in how you think, how you reason about trade-offs, and how you approach complex systems.