AI Security & Platform Engineer

About this job

As an AI Security & Platform Engineer at metaforms in Bengaluru, your day involves testing AI systems adversarially, uncovering vulnerabilities across prompts and tools, and building platform-level controls to keep agent workflows safe by default. Metaforms develops specialized AI infrastructure for research delivery teams, helping market research agencies and sample providers run complex workflows efficiently. In this hands-on technical role, you will write code, inspect production systems, reproduce security issues, and collaborate closely with engineering leadership to implement durable product and infrastructure improvements.

Responsibilities

• Test AI systems adversarially to identify vulnerabilities across prompts, context, data, tools, and execution environments.
• Build platform-level controls, automated evaluations, monitoring, and secure infrastructure.
• Perform threat modeling on new AI features, agent workflows, and tool-calling capabilities before production.
• Review customer-facing applications and APIs for authentication, authorization, and tenant-isolation risks.
• Harden cloud permissions, service identities, storage, networking, and deployment pipelines.
• Add observability for unusual model behavior, tool usage, data access, and token consumption.
• Work directly with engineers to implement production-ready fixes and security regression tests.

Requirements

• 2–3 years of hands-on experience in AI engineering, backend/platform engineering, application security, or infrastructure security.
• Experience building, operating, or securing production AI systems using LLM APIs, agents, or RAG.
• Strong programming ability in Python, TypeScript, Go, or another relevant language.
• Understanding of common web and API security risks like authentication, authorization, and injection.
• Practical experience with cloud infrastructure, CI/CD, containers, or monitoring.
• Ability to investigate vulnerabilities, reproduce them, and contribute effective code fixes.

Good to have

• Hands-on experience with AI red teaming or adversarial model testing.
• Experience securing multi-tenant B2B SaaS products.
• Familiarity with OAuth, webhooks, file processing, and third-party integrations.
• Experience building security test harnesses, fuzzers, or automated evaluations.
• Participation in bug bounties, CTFs, security research, or open-source projects.

Who should apply

This role is ideal for mid-level engineers with two to three years of experience who enjoy combining backend development with security practices for AI platforms. It suits builders who like writing code to solve security challenges, analyzing failure modes in LLMs, and setting up robust infrastructure controls. If you have a strong grasp of software fundamentals and want to secure next-generation AI workflows, you should consider applying.

Frequently asked questions