Security Research Engineer Offensive Research View role →

Cyrion AI; Greater London; Hybrid; GBP 90,000 - 140,000; Full time

Benefits offered by this job

Early equity

Health coverage

$2,000 learning budget per year

Hardware options

Job summary

Cyrion AI is seeking an experienced offensive security engineer to design and extend the exploit-generation and validation pipeline behind our autonomous agent swarm. This hands-on role sits at the center of Cyrion’s offensive capability within a product team, not in isolation.

You will design modules across web, API, and cloud surfaces, build a sandboxed validation pipeline, and reverse-engineer vulnerability classes into repeatable test logic.

Qualifications

• 4+ years in offensive security — pentesting, red teaming, or vulnerability research.
• Comfortable reading and writing exploit code (Python, Go, or similar) against real targets.
• Deep knowledge of at least two of: web/API exploitation, cloud privilege escalation, mobile reversing.
• Experience building or operating tooling that automates parts of an engagement, not just running one manually.
• Clear written communication — you will document findings other engineers rely on.

Responsibilities

• Design and implement new exploit-generation modules across web, API, and cloud attack surfaces.
• Build and maintain the sandboxed validation pipeline that confirms every candidate finding before it ships.
• Reverse-engineer novel vulnerability classes and translate them into repeatable, automatable test logic.
• Partner with the AI/ML team to improve the agents’ hypothesis-generation and attack-chaining reasoning.
• Triage and reproduce edge cases reported by customers and the validation pipeline.
• Publish internal writeups and contribute to the public research blog.

Skills

Offensive security

Exploit development

Python/Go

Web/API exploitation

Cloud privilege escalation

Automation tooling

Technical writing

Job description

Design and extend the exploit-generation and validation pipeline behind Cyrion’s autonomous agent swarm.

You will sit at the center of Cyrion’s offensive capability — the research that decides what our agents are able to find, chain, and prove. This is a hands-on exploit-development role embedded in a product team, not an isolated research function.

What you'll do

• Design and implement new exploit-generation modules across web, API, and cloud attack surfaces.
• Build and maintain the sandboxed validation pipeline that confirms every candidate finding before it ships.
• Reverse-engineer novel vulnerability classes and translate them into repeatable, automatable test logic.
• Partner with the AI/ML team to improve the agents’ hypothesis-generation and attack-chaining reasoning.
• Triage and reproduce edge cases reported by customers and the validation pipeline.
• Publish internal writeups and contribute to the public research blog.

What we're looking for

• 4+ years in offensive security — pentesting, red teaming, or vulnerability research.
• Comfortable reading and writing exploit code (Python, Go, or similar) against real targets.
• Deep knowledge of at least two of: web/API exploitation, cloud privilege escalation, mobile reversing.
• Experience building or operating tooling that automates parts of an engagement, not just running one manually.
• Clear written communication — you will document findings other engineers rely on.

Nice to have

• OSCP, OSWE, or equivalent hands-on certification
• Prior CTF or bug-bounty track record
• Experience with LLM/agentic-AI attack surfaces

Competitive salary plus meaningful early equity, full health coverage, a $2,000/yr learning budget, and your choice of hardware.