AI Agent Defense Engineer Intern
OneSpotAI
Bengaluru, Karnataka, IndiaFresher35m agovia LinkedIn
Posted 16 Nov 2026, 13:18 IST
This job at a glance
- Who can apply
- Freshers and up to 1 year
- Must know
- OperationsPythonLLMsVulnerability managementGitContent writing
- Work
- Remote · internship
- Pay
- Not stated by the employer.
- Skill demand
- LLMs74%Python67%Operations18%Vulnerability management10%of 98 similar fresher jobs posted in India this week
- Listed
- 16 Nov 2026 · still open
Sign in free to see which of these skills your resume already shows.
About the Role
OneSpotAI builds security infrastructure for AI. Companies in healthcare, finance, and logistics rely on AI agents, and we help them harden those agents against misuse. AI agents can be pushed into doing things they were never meant to do. You will find those failure points, understand what causes them, and build the defenses that close them. Your work goes straight into our product.
Key Responsibilities
- Stress-test AI agents for tool misuse, privilege escalation, and data leakage
- Dig into why each failure happened
- Build defenses such as guardrails, permission limits, and detection rules
- Write automated tests so every fix gets verified and stays verified
- Turn findings into short, clear notes engineers can act on
Requirements & Tech Stack
- Strong Python skills
- Hands-on work with LLM apps, agents, RAG, or MCP
- Basic grounding in threat modeling, red teaming, or vulnerability research
- Proof of work: GitHub projects, CTF writeups, blog posts, or experiments
What You Get
- Fully remote, set your own hours
- Stipend at the top of the market for this role
- Work directly with the core team
- Real ownership of what you build