We are an internal Cyber Red Team pioneering the use of Agentic AI to simulate real-world adversaries. This AI-driven role will heavily be dependent on automating real-world attack workflows using Python and LLMs.
We are looking for a strong Python engineer who is curious about security and wants to work on automating offensive techniques using AI. Cyber Security background is NOT mandatory.
You will work directly with experienced Red Team professionals to convert real attack logic into scalable, intelligent systems.
RESPONSIBILITIES:
Build Python-based automation around offensive security tools
Integrate LLM APIs (OpenAI/Anthropic) into real workflows
Design multi-step workflows (input → tool → AI → decision → next step)
Parse and enforce structured outputs (JSON) from AI models
Maintain context/state across tasks and executions
Develop internal APIs and wrappers for tools and services
Continuously improve reliability (error handling, retries, edge cases)
Collaborate with security experts to translate attack techniques into automation
MUST HAVE SKILLS:
Strong Python experience (3–5 years)
Experience in writing robust, production-grade Python. You must be deeply comfortable with Object-Oriented Programming, decorators, context managers, multi threading and async logic.
Write custom Python scripts and API wrappers (such as Model Context Protocol servers)
OOP, clean code, modular design
Comfortable debugging real-world systems
Experience with Docker (containerization)
API & Backend Fundamentals
REST APIs (build and consume)
JSON handling, request/response lifecycle
Hands-on experience with LLM APIs
Gemini/OpenAI
Able to send prompts, parse structured responses
Exposure to LangChain / LangGraph / agent frameworks
Familiarity with deploying code in GCP (Compute Engine or Cloud Run)