Staff AI Engineer, Agentic Automation

Jobgether; United States; Remote; USD 180,000 - 260,000; Full time

Benefits offered by this job

Remote work

Technical ownership

Innovation projects

Architecture leadership

Professional growth

Collaborative environment

Impactful work

Job summary

Jobgether is seeking a senior AI engineering leader to own end-to-end architecture of agentic onboarding and recovery systems. You will design production-ready pipelines, ensure reliability, and lead cross-functional collaboration across engineering teams.

Responsibilities include building LLM-powered agents with tool-use, optimizing performance and cost, and advancing autonomous workflows with strong security practices. Remote-friendly in the United States.

Qualifications

• Shipping LLM-powered agents with tool-use capabilities in production.
• Strong ability to work in Python or TypeScript across environments.
• Deep understanding of agent loops, outputs, and retries.
• Experience building evaluation frameworks or benchmarks.
• Experience with Playwright or similar browser automation tools.
• Knowledge of durable execution and orchestration platforms like Temporal.
• Strong security mindset for credential handling and isolation.
• Ability to lead early-stage systems to production independently.
• Excellent technical writing and cross-team collaboration.
• Hands-on experience with Temporal in production and model serving platforms.

Responsibilities

• Lead AI agent architecture and production ownership for onboarding and self-healing automation.
• Develop agents that analyze API docs and generate OpenAPI specs.
• Create self-healing workflows that diagnose and repair failures before redeploy.
• Establish benchmarks for API specs, browser workflows, and repairs.
• Optimize model accuracy, latency, and cost per output.
• Define security controls for agents handling credentials and untrusted websites.
• Productionize systems with durable workflow orchestration and retries.
• Collaborate across engineering teams and mentor others.
• Drive strategic direction for autonomous onboarding and recovery.

Skills

Production AI Engineering

Python/TypeScript

LLM Agents

Tool Orchestration

Evaluation Frameworks

Browser Automation

Distributed Systems

Security Judgement

Technical Leadership

Technical Writing

Cross-team Collaboration

Temporal in Production

Tools

Temporal

Playwright

Pydantic AI

Fireworks

Together

Amazon Bedrock

AWS

Terraform

Job description

Accountabilities

• AI Agent Architecture and Production Ownership: Lead the end-to-end architecture, development, deployment, and ongoing operation of agentic application onboarding and self-healing automation systems. Define technical roadmaps, establish production-readiness standards, and take ownership of reliability, observability, and on-call responsibilities.
• Intelligent Application Onboarding: Advance AI agents that analyze public API documentation through headless browsers and generate validated OpenAPI specifications. Develop agents that navigate SaaS application interfaces, execute workflows, and produce reliable, replayable automation sequences.
• Self-Healing Automation: Design systems that detect changes in application interfaces, authentication flows, and API structures before they disrupt customer operations. Build agents that diagnose failures, repair workflows or specifications, and validate fixes against real executions before redeployment.
• Evaluation and Quality Assurance: Establish and maintain benchmarks for generated API specifications, browser workflows, and automated repairs. Develop evaluation harnesses and measurable quality thresholds to determine when outputs are safe and reliable enough for production.
• Performance and Cost Optimization: Monitor and optimize model accuracy, latency, and cost per generated output. Evaluate frontier and open-weight models using measurable performance data to balance quality, efficiency, and scalability.
• Security and Risk Management: Define and implement security controls for AI agents interacting with sensitive credentials and untrusted websites. Address prompt injection, credential isolation, token management, sandboxed browsing, and network egress controls. Ensure secrets are excluded from prompts, logs, queues, and persisted workflow state.
• Distributed Systems and Workflow Reliability: Productionize agent systems using durable workflow orchestration technologies, such as Temporal, and implement robust handling of retries, idempotency, asynchronous execution, and failures.
• Cross-Functional Technical Leadership: Collaborate with integration, automation-runtime, and platform engineering teams to align architecture and implementation. Produce clear technical design documentation, guide engineering decisions, and mentor contributors as agentic automation capabilities mature.
• Continuous Improvement and Strategic Direction: Drive the evolution of agentic automation toward increasingly autonomous application onboarding and recovery. Establish milestones for production adoption, reduced failure rates, lower operating costs, and minimizing the need for manual intervention.

Requirements

• Production AI Engineering Experience: Demonstrated experience shipping LLM-powered agents with tool-use capabilities that operate autonomously for real users, alongside ownership of backend systems in production.
• Programming Skills: Strong proficiency in Python or TypeScript, with the ability to work productively in both languages.
• LLM Agents and Tool Use: Deep understanding of agent loops, structured outputs, tool orchestration, retries, and common failure modes in autonomous AI systems.
• Evaluation and Benchmarking: Experience building evaluation frameworks, benchmarks, or testing harnesses and using quantitative results to guide technical decisions and improve production quality.
• Browser Automation and API Integrations: Significant experience with browser automation technologies such as Playwright or comparable tools, or deep expertise in backend and API integration systems, combined with the ability to work effectively across both areas.
• Distributed Systems: Understanding of queues, idempotency, durable execution, and workflow orchestration platforms such as Temporal, AWS Step Functions, or similar technologies.
• Security Expertise: Strong security judgment when designing systems that handle credentials and sensitive information, including familiarity with prompt injection risks, secret isolation, and secure execution environments.
• Technical Leadership: Ability to take early-stage systems from development to production independently, make sound architectural trade-offs, and ensure systems remain reliable after launch.
• Communication and Collaboration: Excellent technical writing skills, with the ability to document architecture, communicate complex decisions, influence stakeholders, and mentor engineers across teams.
• Preferred Technical Experience: Hands-on experience with Temporal in production, Pydantic AI or comparable agent frameworks, open-weight model serving platforms such as Fireworks, Together, or Amazon Bedrock, and LLM cost modeling.
• Additional Preferred Skills: Familiarity with computer-use agents, Chrome Manifest V3 browser extensions, OpenAPI tooling, identity and access management (IAM), credential management, AWS, and Terraform.

Benefits

• Remote work: Opportunity to work remotely from anywhere in the United States.
• Technical ownership: Lead critical AI automation systems from early development through production and ongoing operation.
• Innovation: Work on challenging problems at the intersection of agentic AI, enterprise security, browser automation, and identity management.
• Technical influence: Shape architecture, engineering standards, evaluation frameworks, and the long-term direction of agentic automation.
• Professional growth: Develop expertise in production-grade LLM systems, distributed workflows, self-healing automation, and AI security.
• Collaborative environment: Partner with engineering specialists across integrations, platform development, and automation infrastructure.
• Meaningful impact: Help enterprises reduce manual identity-management work, improve automation reliability, and strengthen security across disconnected applications.