Staff AI Engineer, Agentic Automation

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

📋 Description

• Oversee the complete onboarding process for agentic systems and self-healing, which includes architecture design, roadmap development, and on-call support post-launch.

• Transition existing API catalog generation and browser workflow generation systems into production with durable workflows utilizing Temporal, ensuring proper observability.

• Identify automation drift from failed processes before it impacts customers.

• Develop agents capable of diagnosing failures, repairing workflows or specifications, and re-validating solutions against actual runs.

• Determine which issues can be resolved automatically and which ones necessitate human intervention.

• Establish benchmarks for generated specifications, workflows, and repairs.

• Set defined quality thresholds to assess production readiness.

• Manage cost and latency for each output, selecting between frontier and open-weight models based on accuracy and cost metrics.

• Define the threat model, addressing concerns such as prompt injection from page content, credential isolation, token management, sandboxed browsing, and egress.

• Ensure that credentials are excluded from prompts, logs, queues, and stored run states.

• Collaborate with integration, automation-runtime, and platform teams.

• Create design documents and provide mentorship to contributors.

⛳️ Requirements

• Proven experience in deploying an LLM agent with tool utilization that operated unattended for real users.

• Demonstrated ownership of a backend system in a live production environment.

• Expertise in Python or TypeScript, with proficiency in both languages.

• Familiarity with LLM agents and tool usage, encompassing agent loops, structured outputs, retries, and failure modes.

• Experience in creating benchmarks or evaluation harnesses and applying them for informed decision-making.

• In-depth knowledge of browser automation (such as Playwright or similar) or backend and API-integration systems, with comfort in the alternative domain.

• Understanding of distributed systems, including queues, idempotency, and durable workflow engines (like Temporal, Step Functions, or similar).

• Sound security judgment for systems managing secrets, particularly concerning prompt injection risks.

• Strong technical writing skills with the ability to articulate trade-offs that other teams can rely on.

• Preferred: Experience with Temporal in a production environment; familiarity with Pydantic AI or a comparable agent framework.

• Preferred: Knowledge of open-weight model serving (Fireworks, Together, Bedrock) and LLM cost modeling.

• Preferred: Experience with computer-use agents or browser extensions (Chrome MV3); familiarity with OpenAPI tools.

• Preferred: Background in identity, IAM, or credential management; experience with AWS and Terraform.

🏝️ Benefits

• Competitive salary and performance-based bonuses.

• Flexible working hours and remote work options.

• Opportunities for professional development and continuous learning.

• Comprehensive health benefits including medical, dental, and vision coverage.

• Collaborative and inclusive company culture.