AI Engineer

This is a fully remote position, open to applicants in United States.

📋 Description

• Construct and enhance the foundation of Nava’s LLM-driven products.

• Develop reusable components for context management, agent state, sandboxed execution, and model routing applicable across various AI scenarios.

• Design deployment and state management strategies that ensure the continuity of long-running agent tasks during updates, restarts, and retries.

• Facilitate the resumption, inspection, and replay of agent tasks.

• Package context into secure execution environments across different platforms.

• Eject and retrieve fresh context as agents operate.

• Direct models to suitable tasks and coordinate subagents effectively.

• Assess and optimize token utilization while minimizing costs per task as usage scales.

• Mitigate unplanned platform interruptions that affect AI product teams.

• Document, test, trace, and evaluate shared platform methodologies.

• Enable new engineers to swiftly adopt shared practices and identify regressions before they impact users.

⛳️ Requirements

• Direct ownership of infrastructure supporting an LLM-based product relied upon by other engineers or users.

• Proficiency in agent runtime, context or state management, orchestration, or evaluation and tracing.

• Experience in deploying LLM features to production while managing model behavior, latency, costs, and trade-offs.

• Modern full-stack development expertise in TypeScript and/or Python.

• Experience incorporating foundation-model APIs, tool invocation, and human review into applications.

• Practical skills in evaluation and tracing.

• Capability to articulate decisions to product, design, and domain specialists.

• Ability to transform product and domain challenges into reusable functionalities.

• Familiarity with React, Node, TypeScript, Python, Postgres, and AWS as relevant to the technology stack.

• Experience with various foundation-model providers.

• Knowledge of sandboxed execution environments for agents.

• Utilization of Langfuse for tracing and evaluation within the stack.

• No specific framework, model provider, or minimum audience size requirements.

• Coding agents are encouraged but not mandatory during the interview process.

🏝️ Benefits

• A dedicated environment to build the AI platform.

• A remote-first work culture.

• Freedom to excel in your role.

• Access to cutting-edge technology and tools.

• A collaborative team atmosphere.

• Direct communication with leaders influencing product and platform decisions.

• Preparation notes provided before each stage of the interview process.