Cyber-summer at the Cybrela academy. 6 cybersecurity webinars, online every other Friday. Register for free ➡️

AI Engineer / Builder – Internal AI Tools

  • Hybrid (Prague 7 / online)
  • Freelance (part-time / project-based) or full-time employment

Cybrela AI/tech is the technology arm of the Cybrela consulting group (cybersecurity and compliance – NIS2, DORA, ISO 27001, CRA). We're building a cyber-compliance AI tool and looking for an engineer to join the team who will build it from scratch into a functional, production-ready state and gradually help with other internal AI tools and automations across the company. It's a greenfield project. You'll largely design the architecture and the way of working yourself. This isn't just about "getting it running" – it's about getting it running securely, bearing in mind that we work with sensitive areas of cybersecurity and compliance.

What you'll do

  • RAG pipeline: ingest (Teams / video / documents) → transcription (Whisper) → anonymization / pseudonymization → embeddings → search.
  • LLM: main inference via AWS Bedrock (eu-central-1), local models (Ollama / Mistral) for sensitive data.
  • Fine-tune response quality and stability.
  • Solve PII pseudonymization from transcripts and integrate web search.
  • Handle deployment and operations (Docker, AWS, monitoring) – making it run, not just a demo.
  • Gradually help with other internal tools and automations.

What we expect from you

MUST HAVE

  • Strong Python and experience building and operating real applications (not just scripts / notebooks).
  • Hands-on experience with LLM applications: RAG, embeddings, vector DB, prompt engineering, quality evaluation.
  • Experience with LLM APIs and local models (Anthropic / Bedrock + Ollama / Mistral / vLLM, etc.).
  • Data pipelines: text processing, audio transcription (Whisper), data cleaning and anonymization.
  • AWS and standard deployment (Docker, networking, S3, compute) – the ability to take a system all the way to production.
  • Fluent use of Claude Code / agentic coding tools.
  • Git, basic CI/CD.
  • A strong sense for security and data protection.

NICE TO HAVE

  • Front-end for internal UI (React / Next, or Streamlit / Gradio for quick prototypes).
  • MLOps / observability (Langfuse, eval harness, logging).
  • FastAPI / backend API.
  • Experience in a regulated / legal-tech / compliance environment.
  • Familiarity with the EU AI Act and data governance

What we offer

  • Ownership of the project from scratch – you design the architecture, you're not cleaning up after anyone.
  • Variety: the product core as well as internal automation, not one ticket after another.
  • Fast decision-making, no corporate fluff.
  • A meaningful domain (security and compliance) and real impact.

Interested in the role?

Let us know about yourself – send us your CV or a link to your LinkedIn. We'll get back to you within a few days.
Send CV