Agentic Engineer

About the Role

This is not a prompt engineering role. We are looking for engineers who can build robust, reliable, and autonomous systems. You will be defining the cognitive architectures of the future—building agents that can plan, reason, code, and execute complex workflows without human intervention. You will work at the bleeding edge of applied AI, solving unsolved problems in context management, tool use, and multi-agent orchestration.

What You'll Do

• Architect multi-agent systems that can solve open-ended problems.
• Build evaluation frameworks to measure and improve agent reliability.
• Optimize latency and cost while maximizing reasoning capabilities.
• Develop novel techniques for memory management and long-horizon planning.
• Push the boundaries of what is possible with current foundation models.

Requirements

• Deep intuition for LLM behavior, limitations, and prompting strategies.
• Strong software engineering fundamentals (Python/TypeScript) to build reliable scaffolding around stochastic models.
• Experience building complex RAG pipelines and agentic workflows (LangGraph, AutoGen).
• Ability to evaluate agents using rigorous benchmarks (evals are unit tests for AI).
• Comfort reading and implementing research papers (arXiv is your documentation).

The Interview

This track requires just one 30-minute interview. Here's what you can expect:

• Resume Deep Dive: Detailed walkthrough of your past experience.
• Technical Case Study: Short case study on a technical topic.
• Operating Case Study: Short case study on an operating topic.

Frequently Asked Questions