Senior AI Software Engineer, AI Enablement

Extend
•$155,000 - $170,000

About The Position

The Senior AI Software Engineer will join Extend’s AI Enablement team, which equips teams across the company with AI tools. The role works directly with business users in areas such as finance, support, claims, sales, recruiting, and operations to understand workflows, identify useful opportunities, and build reliable engineering solutions. It will develop the platform and practices that turn prototypes into durable, secure, measurable tools, including MCP connectors, agent-to-agent integrations, shared agent skills, and evaluation systems. The engineer will also help establish security, governance, observability, and adoption practices for AI agents and skills. This includes credential scoping, risk-tiering, sandboxing, dependency scanning, governance reviews, end-to-end instrumentation, documentation, onboarding, and mentoring junior engineers. The position is based at Extend, whose headquarters is in downtown San Francisco; the posting does not specify the work arrangement for this role.

Requirements

  • At least 5 years of experience shipping production systems.
  • Proficiency in TypeScript/Node, Python, or Rust.
  • Hands-on experience building with MCP and at least one agent framework or harness, such as Claude Code, OpenAI Agents SDK, LangChain/LangGraph, CrewAI, AutoGen, or Google ADK.
  • Experience designing and running evaluations for LLM or agent systems and using results to guide shipping decisions.
  • Experience building third-party integrations, including OAuth flows, credential scoping, webhooks, pagination, rate limiting, and retries.
  • Experience building developer tooling, platforms, or internal tools relied upon by other people, with adoption used as a measure of impact.
  • Ability to work directly with non-technical stakeholders, facilitate working sessions, challenge assumptions constructively, and explain trade-offs in plain language.
  • Comfort with AWS, including Lambda, API Gateway, DynamoDB, and S3, and solid security fundamentals, including IAM, KMS, and serverless attack patterns.
  • Familiarity with AI governance and legal or privacy review processes, including risks related to data access, permissions, tool use, and autonomous actions.
  • Regular use of AI tooling and informed opinions about what works.

Nice To Haves

  • Familiarity with evaluation and observability tooling such as LangSmith, Langfuse, Braintrust, OpenTelemetry, Coralogix, or Datadog.
  • Knowledge of LLM application security, including the OWASP Top 10 for LLMs and agentic applications, prompt injection defense, and agent sandboxing.
  • Experience with AWS CDK, CloudFormation, or similar infrastructure-as-code tools.
  • Background in platform engineering, internal developer platforms, or enterprise integration.
  • Prior individual-contributor experience at an early-stage startup with a small team and broad responsibilities.

Responsibilities

  • Work with business users to analyze workflows, map processes, identify high-impact problems, and turn ambiguous requests into scoped technical plans.
  • Design and deliver secure MCP connectors to internal systems and third-party SaaS platforms, including finance, CRM, data warehouse, expense management, product analytics, support, and ATS systems.
  • Build agent-to-agent integrations using A2A and similar protocols so specialized agents can delegate safely across domains.
  • Select and configure agent harnesses or orchestrators and build multi-agent patterns where appropriate.
  • Establish agent and skill evaluation practices, including golden datasets, regression suites, LLM-as-judge, tracing, and drift detection.
  • Build and maintain a shared library of reusable agent skills, codify patterns, and promote safe, high-quality skill development.
  • Design credential scoping and vending for agent connectors, using per-user, per-skill, and per-connector keys, OAuth/OIDC where appropriate, and least privilege.
  • Develop a risk-tier and review model for shared skills, including sandboxing and dependency scanning for skills that use untrusted packages.
  • Partner with legal, security, and privacy teams on governance reviews for agents that access sensitive data or take autonomous actions.
  • Instrument the platform with OpenTelemetry so MCP calls, skill executions, and credential use are visible in Coralogix.
  • Drive adoption by writing self-service guides, onboarding teams, and incorporating user feedback; help teams become self-sufficient.
  • Mentor junior engineers in evaluations, agent design, and secure integration work.

Benefits

  • Full medical, dental, and vision benefits.
  • Stock in an early-stage startup.
  • Generous, flexible paid time off policy.
  • 401(k) with financial guidance from Morgan Stanley.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service