Staff Software Engineer, Frontier Security Team

Overview

Staff Software Engineer for the Frontier Security AI team, leading development of an Agentic Harness and agent evaluation platform. The role spans AI agents, distributed infrastructure, evaluation systems, security, and production operations.

What you'll do

• Architect and build an Agentic Harness for complex, multi-step AI workflows.
• Design interfaces for tool execution, context, state, memory, permissions, retries, and human review.
• Build evaluation harnesses, datasets, automated graders, experiment pipelines, and release gates.
• Analyze production agent trajectories and convert failures into tests and fixes.
• Develop measurements for task completion, correctness, safety, latency, reliability, and cost.
• Productionize secure, observable, multi-tenant model capabilities.
• Set technical direction, lead cross-team projects, and mentor engineers.

What you'll need

• 9+ years of software engineering experience, including technical leadership of complex production systems.
• Experience shipping and operating LLM applications, AI agents, or model-backed workflows in production.
• Strong distributed systems, service architecture, API, concurrency, and failure-handling experience.
• Fluency in Python and proficiency in Java, Go, Rust, or TypeScript.
• Knowledge of tool calling, structured generation, retrieval, context engineering, prompt management, and model APIs.
• Bachelor's degree in Computer Science, Engineering, or a related field, or equivalent practical experience.
• Clear written and verbal communication with engineering, product, and leadership audiences.

Nice to have

• Experience building evaluation or observability infrastructure for agentic systems.
• Experience with human evaluation, multi-agent orchestration, simulations, adversarial tests, or safety guardrails.
• Experience with retrieval systems, multi-tenant sensitive-data systems, model training, or frontier-model evaluation.

Details

• Location: Bangalore, India.