Full-Stack Developer AI-Native

About the role (AI-Native Engineering · Remote / Hybrid)At Imperum, we are building an agentic AI platform for security operations – an autonomous SOC that detects, investigates, contains, and remediates threats with minimal human intervention. Our product is not “AI-assisted” security; it is a system of AI agents that run the SOC loop end-to-end, from alert triage to incident response and forensic analysis. We are looking for a Full-Stack Developer who brings both modern full-stack engineering and real security operations domain knowledge. You will ship features across the platform while using AI coding tools as a core part of your own development loop. You must be fluent in agentic AI patterns – tool use, multi-agent orchestration, evaluations, and guardrails – because that is the fabric of the product itself. And you must understand how SOC, incident response, and forensics actually work, because you are building for analysts who will trust our agents with real investigations.What you’ll do•    Build and ship full-stack features across the frontend, backend, APIs, and data layer, using AI coding tools as your primary force multiplier.•    Translate product requirements into clear specs, plans, and prompts that AI agents can execute reliably — and iterate on them until the output is production-grade.•    Own the quality bar: review, refactor, and harden AI-generated code. You are the engineer of record for everything that lands in main.•    Design and evolve the architecture and conventions documentation (agent context, coding standards, project rules) so that AI assistants consistently produce code that fits our codebase.•    Integrate AI into the day-to-day workflow: scaffolding, test generation, refactors, migrations, code review, documentation, and incident triage.•    Work with MCP servers, custom tools, and internal agents to automate repetitive engineering tasks and extend what AI assistants can do in our environment.•    Contribute to observability, CI/CD, and deployment pipelines so that AI-authored changes ship safely and are easy to roll back.•    Pair with product, design, and other engineers to ship end-to-end features — AI is a teammate, not a replacement for collaboration.What we’re looking forMust-have•    EU nationality (or existing right to work in the EU without sponsorship). We are not able to sponsor work visas or relocation from outside the EU for this role.•    Minimum 5 years of professional full-stack experience shipping production software.•    Strong hands-on experience with our core stack: Python and Go on the backend, React on the frontend. You don’t need to be equally expert in all three, but you should be comfortable working across them.• Proven knowledge of security operations – SOC, Incident Response, and Forensics – is a must. You have worked in, built for, or worked closely with SOC / DFIR teams. You understand the alert triage and investigation lifecycle, common detection logic (SIEM rules, EDR telemetry, MITRE ATT&CK), the IR lifecycle (detect, contain, eradicate, recover, lessons learned), and forensic fundamentals (evidence handling, chain of custody, host/network/memory artifacts, timeline analysis). You can tell a good agent-driven investigation from a bad one because you’ve done the work yourself.• Agentic AI as core knowledge. You have built, integrated, or shipped features powered by LLM agents – not just called a chat API. You are comfortable with tool use / function calling, multi-step planning and multi-agent orchestration, RAG, MCP, evaluations, guardrails, and the failure modes of autonomous agents (hallucinations, loops, prompt injection, unsafe tool calls). You know how to design agents that are reliable enough to trust in production.•    Working knowledge of relational and/or document databases (PostgreSQL, MongoDB, or similar) and basic data modeling.•    Comfortable with Git, code review, testing, and CI/CD as non-negotiable practices.•    Hands-on, regular use of AI coding assistants (Claude Code, Cursor, Copilot, Windsurf, Aider, or equivalent) on real projects — not just demos.•    Strong prompt and context engineering skills: you know how to scope tasks, feed the right context, and iterate with AI agents to get reliable results.•    A critical eye for AI output: able to spot hallucinated APIs, subtle bugs, insecure patterns, and over-engineered code, and fix them fast.•    Fluent English (spoken and written) is a must – it is our working language across code, docs, meetings, and async communication. You can turn vague ideas into clear specs, ADRs, and prompts that both humans and AI agents can act on.Nice-to-have•    Experience building or integrating security tooling: SIEM (Splunk, Elastic, Sentinel), EDR/XDR, SOAR platforms, threat intelligence feeds, or detection engineering.•    Familiarity with event-driven or workflow systems (Kafka, Temporal, queues) and container-based deployment (Docker, Kubernetes).•    Certifications or hands-on background in offensive/defensive security (GCIH, GCFA, OSCP, CEH, Blue Team labs) or a track record contributing to detection rules, IR playbooks, or threat research.•    DevOps / platform experience: infrastructure as code, observability, or security hardening.•    Open-source contributions, technical writing, or a track record of sharing how you work with AI tools.Our cadence: idea to release in 5 daysWe operate on a strict 5-day delivery cycle: from the moment an idea is greenlit to the moment it ships to production, we give ourselves one working week. No exceptions, no quietly sliding deadlines. This cadence is only possible because AI coding tools are deeply embedded in how we build, and because every engineer on the team is fluent in using them.In practice, a cycle typically looks like this:• Day 1 – Spec & shape. Scope the idea, write the spec and the agent context (project rules, prompts, constraints), and break the work into tasks an AI agent can execute reliably.• Days 2–3 – Build. Implement end-to-end with AI agents, iterating fast across frontend, backend, and data. Tests are written alongside the code, not after.• Day 4 – Harden. Review AI-generated code in depth, refactor, tighten security and observability, and run it against real data and edge cases.• Day 5 – Ship. Deploy to production, monitor, document, and close the loop. What doesn’t fit in the cycle is cut or descoped – not delayed.This rhythm is the job. If the idea of shipping a meaningful feature every week sounds energizing rather than terrifying, you’ll fit right in.How we work•    Our stack: Python and Go for backend services, React for the frontend. We pick the right tool per problem, but these are our defaults.•    AI-first, not AI-only: every engineer is accountable for the code that ships. AI accelerates us; it does not excuse us.•    Small, senior team. Short feedback loops. High autonomy and high ownership.•    We invest in tooling: prompt libraries, shared conventions, custom MCP servers, and internal agents that make everyone faster.•    We pay for the best AI tools and models so you always have the right tool for the job.•    Async-friendly, remote-first, with regular in-person time when it matters.What we offer•    Compensation package combining competitive base salary, performance bonus, and RSUs (Restricted Stock Units).•    Top-tier hardware and a budget for AI tools, courses, and conferences.•    Flexible remote or hybrid setup.•    A team that treats AI-assisted engineering as a craft worth mastering — you will grow fast here.How to applySend us your CV, your GitHub or portfolio, and a short note on how you use AI coding tools in your day-to-day work. If you have a favorite prompt, workflow, or war story about shipping something non-trivial with an AI agent, we want to hear it.