Skip To Main Content
backBack to Search

Senior Penetration Testing Engineer

Remote in Ukraine: Lviv
Security.Testing& 15 others
hot
Looking for something else?

Find a vacancy that works for you. Send us your CV to receive a personalized offer.

Find me a job

We are seeking a Senior Penetration Testing Engineer to develop exploits for newly discovered vulnerabilities and attack LLM applications and autonomous agents as a capable adversary would. This is a research-first role with no fixed checklist: you will decide what to investigate, dig into new attack techniques as they appear and turn what you learn into findings, exploits and tooling, contributing original research to the client's zero-day program and helping build the AI-driven tooling that scales its offensive capability.

Responsibilities
  • Run vulnerability research on binaries, firmware, IoT devices, web applications and enterprise software, using reverse engineering, code review and fuzzing
  • Track new CVEs, advisories, papers and conference talks, analyze patches and emerging attack techniques, especially in AI/LLM security, and decide which targets and ideas are worth pursuing
  • Write reliable proof-of-concept exploits for memory corruption, logic flaws and authentication bypasses
  • Test LLM applications and agentic systems end-to-end, covering prompt injection, RAG poisoning, tool and permission abuse, and data exfiltration
  • Map findings to OWASP Top 10 for LLM Applications and MITRE ATLAS
  • Build and improve AI-driven offensive tooling, such as automated fuzz harness generation, triage and agentic testing frameworks
  • Deliver client engagements and explain findings to both engineers and executives
Requirements
  • 5+ years of experience in offensive security, with deep experience in vulnerability research and exploit development
  • Proof of original work such as CVE credits, published advisories or exploits, conference talks, or strong bug bounty/CTF results
  • Hands-on experience testing LLM or agentic AI systems and understanding of how models, retrieval, tools and permissions create attack surface
  • Proficiency in Python and comfort with C/C++ and x86/x64 or ARM assembly
  • Experience with a modern RE toolchain such as IDA, Ghidra or Binary Ninja, a debugger, and a fuzzer such as AFL++ or libFuzzer
  • A self-driven mindset, finding your own leads, learning new domains quickly, and being comfortable working with little guidance
Nice to have
  • Skills in Rust or Go
  • Experience with Garak, PyRIT or Promptfoo
  • OSEE, OSED or OSCE3