Build Agent Runtime Systems; Develop and improve agent runtimes
Design tool orchestration frameworks
Build memory and context management systems
Implement verification loops and agent self-checking mechanisms
Improve reliability and execution accuracy
Ship Integrations; Build integrations with business platforms and APIs
Implement OAuth, webhooks, schema mapping, and data synchronization
Create reliable workflows that operate on real customer data
Support integrations across platforms such as: Salesforce, HubSpot, Stripe, Shopify, QuickBooks, Notion, Microsoft Teams, Slack
Manage containerized workloads
Improve scalability and system reliability
Optimize infrastructure performance and operational costs
Support high-volume production workloads
Ship user-facing features directly into production
Collaborate directly with founders and product stakeholders
Own technical surfaces end-to-end

Experience building or significantly modifying AI agents
Strong Python development skills
Strong TypeScript development skills
Experience with agent architectures and orchestration systems
Experience shipping production software rapidly
Strong systems thinking and architectural decision-making abilities
Experience working with APIs, integrations, and distributed systems
Strong understanding of AI-assisted development workflows
Open-source contributions within AI ecosystems (preferred)
Experience with developer tooling and automation (preferred)
Experience creating custom agent workflows, MCP servers, or CLI tooling (preferred)
Infrastructure and DevOps experience (preferred)
Experience with Slack or Microsoft Teams applications (preferred)
Previous startup or founder experience (preferred)
Deep interest in AI systems, evaluation methodologies, and emerging models (preferred)

Mid-Senior Agent Harness Engineer, Python, TypeScript

Key skills