Core AI Engineer — Distributed Systems & Evaluation

Sycamore; Palo Alto (CA); On-site; USD 180,000 - 250,000; Full time

Benefits offered by this job

Competitive cash compensation

Equity

Job summary

Sycamore is building an enterprise-grade agent operating system to orchestrate agentic apps with strong security and control. This role focuses on backend, AI-driven learning data planes, evaluation systems, and durable orchestration for long-running tasks in a fast-moving startup environment.

You will design and implement the end-to-end pipelines that power agent learning, evaluation, and safe self-improvement while shaping product-facing features in React/TypeScript.

Qualifications

• 5-12 years of software engineering experience, with exceptions accepted for exceptional candidates.
• Strong backend and distributed-systems fundamentals including typed API design, async workflows, persistence and production debugging.
• Depth in at least one area, with interest in others.
• Empirical approach to AI quality: form hypotheses, design evaluations, interpret data.
• Ability to reason about retries, idempotency, partial failure, long-running state, concurrency, latency, and cost.
• Security-minded approach to multi-tenant systems, identity, authorization, credentials, privacy, and auditability.
• Product judgment: durable abstractions behind real requirements without over-generalizing.
• AI-native mindset using models as a force multiplier while owning architecture and outcomes.
• Comfort building product-facing software in React and TypeScript; handle streaming state and accessibility.
• Comfort with startup ambiguity, fast feedback loops, and broad ownership.

Responsibilities

• Develop learning data plane around agents: structured trajectories, feedback signals, offline datasets, lineage and privacy controls.
• Create evaluation systems for task completion, tool use, long-horizon behavior, safety, latency and cost, and validate numbers.
• Design experiment and versioning systems for replay, canaries, and controlled rollouts with clear promotion and rollback criteria.
• Build durable orchestration for long-running tasks, checkpoints, handoffs, and human-in-the-loop interactions with typed tool interfaces and memory retrieval enforcing boundaries.
• Own execution environments for agents including isolation, startup performance, resource limits, cost, and the policy layer for staged changes.
• Publish reliable APIs, event-driven interfaces, and reusable libraries across agent categories and deployments.

Skills

Software engineering

Distributed systems

Backend design

AI evaluation

Retry & idempotency

Security & multi-tenant

Product judgment

AI-native

React/TypeScript frontend

Startup mindset

Job description

Sycamore is building an enterprise-grade agent operating system to orchestrate agentic apps with strong security and control. This role focuses on backend, AI-driven learning data planes, evaluation systems, and durable orchestration for long-running tasks in a fast-moving startup environment.

You will design and implement the end-to-end pipelines that power agent learning, evaluation, and safe self-improvement while shaping product-facing features in React/TypeScript.