AI Engineer

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

📋 Description

• Develop the runtime for agents specific to various roles, incorporating tool usage, delegation, pause/resume functionalities, durable checkpoints, retries, and failure recovery mechanisms.

• Establish deterministic controls on non-deterministic reasoning, including governed reads, suggested writes, precondition checks, business invariants, scoped permissions, idempotency, audit trails, and rollback capabilities.

• Create tenant-scoped context assembly, retrieval processes, freshness controls, provenance tracking, transcripts, artifacts, tool outputs, and evidence that is replayable.

• Implement human-in-the-loop gates, approval thresholds, reversibility, enforcement of tenant policies, and audit history for sensitive workflows.

• Conduct both offline and online evaluations, develop scenario libraries, simulations, trajectory reviews, regression detection, telemetry for cost and latency, and gates for autonomy promotion.

• Assist product teams in packaging prompts, tools, context requirements, policies, evaluations, rollout controls, ownership, and rollback into governed capabilities.

• Make architectural tradeoffs regarding models and inferences, considering latency, costs, quality, structured outputs, caching, fallback behavior, provider selection, and model routing.

• Design and implement essential services for the Agent OS platform.

• Write production-level code and review implementation details from fellow engineers.

• Create reliable APIs, workflows, tools, and services for agent operations.

• Monitor traces, troubleshoot failures, and enhance production performance.

• Design evaluation scenarios and regression test suites for agent workflows.

• Collaborate with domain teams to transform agent use cases into reusable platform patterns.

• Contribute to the technical direction while ensuring a focus on rapid and safe shipping.

• Work together with Product, Architecture, Security, Data Platform, Atlas, and domain engineering teams.

⛳️ Requirements

• Proven skills and experience in developing production-grade AI agent platforms.

• Capability to construct agent runtimes, typed tools, context and memory systems, trust and approval flows, evaluation infrastructure, and production observability.

• Proficiency in designing reliable APIs, workflows, tools, and services for agent execution.

• Competence in inspecting traces, debugging failures, and enhancing production behavior.

• Ability to create evaluation scenarios and regression test suites for agent workflows.

• Skills to address stale context, incorrect tool calls, missing permissions, unsafe actions, retrieval issues, latency spikes, and cost regressions.

• Ability to define platform contracts for tools, actions, approvals, context, memory, evidence, and evaluations.

• Capability to make trade-offs regarding latency, cost, quality, structured outputs, caching, fallback behavior, provider choice, and model routing.

• Strong communication skills to engage effectively with engineers, product managers, architects, security partners, and engineering leadership.

🏝️ Benefits

• Flexible time off.

• Opportunities for learning and development.

• Comprehensive onboarding program.

• Leadership training.

• Various programs and events.

• Bonusly rewards.

• Peer-nominated awards.

• Company-covered medical, dental, and vision insurance.

• 100% employer-paid options.

• 90% coverage for dependents.

• Flexible Spending Account (FSA) and Health Savings Account (HSA).

• 401k matching.

• Telehealth options including memberships with One Medical.

• Parental leave and support services.

• Up to $20,000 in fertility services, including IUI and IVF.

• Reimbursement for surrogacy.

• Adoption reimbursement.

• On-demand maternity support through Maven Maternity.

• Complimentary breast milk shipping through Maven Milk.

• Pet insurance.

• Legal advisory services.

• Financial planning tools.

• Annual bonus opportunities.

• Equity options.