Astra Security
Website:
getastra.com
Job details:
About Astra:
Astra is a cyber security SaaS company that makes otherwise chaotic pentests a breeze with its one of a kind Pentest Platform. Astra's continuous vulnerability scanner emulates hacker behavior to scan applications for 9300+ security tests. CTOs & CISOs love Astra because it helps them fix vulnerabilities in record time and move from DevOps to DevSecOps with Astra's CI/CD integrations.
Astra is loved by 650+ companies across the globe. In 2023 Astra uncovered 2 million+ vulnerabilities for its customers, saving customers $69M+ in potential losses due to security vulnerabilities.
We've been awarded by the President of France Mr. François Hollande at the La French Tech program and Prime Minister of India Shri Narendra Modi at the Global Conference on Cyber Security. Loom, MamaEarth, Muthoot Finance, Canara Robeco, ScripBox etc. are a few of Astra’s customers.
With over 18,000 vulnerabilities detected daily, Astra is building the next generation of autonomous security testing.
Role Overview
A frontier model alone cannot reliably perform penetration testing. The real challenge lies in building the harness around it: orchestrating models and tools, validating findings, evaluating accuracy, and optimizing cost and performance.
As an AI Engineer, Autonomous Pentesting, you will build and productionize AI-powered security testing systems, combining LLM engineering, backend development, and security evaluation to deliver reliable autonomous pentesting at scale.
What You'll Do
- Build AI agents: Develop and productionize autonomous pentesting workflows across web applications, APIs, mobile, and cloud.
- Engineer LLM workflows: Implement prompting, tool calling, agent orchestration, retrieval, and context management.
- Optimize models: Orchestrate frontier and open-weight models, exploring distillation and fine-tuning to improve cost and performance.
- Build backend systems: Develop scalable REST APIs, asynchronous workflows, distributed services, and efficient database architectures.
- Own evaluation: Build ground-truth datasets and measure precision, recall, false positives, run-to-run variance, latency, and cost per true finding.
- Ensure reliability: Implement validation, guardrails, monitoring, and error handling to prevent incorrect findings from reaching customers.
- Ship end-to-end: Take solutions from prototype to production, collaborating with engineers, pentesters, and security researchers.
What We're Looking For
- 3- 6 years of experience in backend engineering, AI engineering, or production software development.
- Strong proficiency in Python or another programming language such as Java, C++, C#, or PHP.
- Hands-on experience building LLM-powered applications, agents, or model orchestration workflows.
- Experience with REST APIs, databases, asynchronous processing, and distributed systems.
- Strong software engineering fundamentals, debugging skills, and production ownership.
- Ability to evaluate AI outputs rigorously using ground truth, measurable benchmarks, and false-positive analysis.
- Sufficient security understanding to distinguish genuine vulnerabilities from incorrect findings.
Nice to Have
- Experience in penetration testing, offensive security, CTFs, or vulnerability research.
- Experience with open-weight model serving, fine-tuning, or distillation.
- Familiarity with mobile security, static/dynamic analysis, or reverse engineering.
- Experience taking research prototypes into production.
Why Astra?
- Build core AI technology for autonomous penetration testing.
- Own projects from experimentation to production.
- Solve challenging problems across AI, cybersecurity, and distributed systems.
- Work closely with founders, engineers, and security researchers.
- Enjoy a hybrid work environment, health insurance, and team benefits.
Click on Apply to know more.