Senior Software Engineer II - Agentic Intelligence

Honeycomb.io
•$183,340 - $206,000•Remote

About The Position

Honeycomb is hiring a Senior Software Engineer II for its Agentic Intelligence team. The team builds AI agents that investigate, reason over, and act on real observability data in Canvas, Honeycomb’s agentic workspace for engineers. Its work includes Canvas, the Honeycomb MCP server, and Canvas Agent and Canvas Skills surfaces. This role will bring deep agent engineering expertise to expand those capabilities, including new agents, memory, spatial awareness, and improvements to the Bedrock loop. The role focuses on building production-grade agents that use Honeycomb’s fast, high-cardinality data store to investigate systems in ways conventional observability backends cannot. Work may begin as a prototype, but the engineer is expected to carry it through to a durable production implementation. The position is agent-focused and also includes some full-stack development, along with ownership of agent evaluation, maintenance, and product direction. Honeycomb is a fully distributed company that emphasizes impact, inclusion, trust, autonomy, and accountability. The company describes its culture as distributed-first and provides a home office, co-working, and internet stipend. Honeycomb cannot currently sponsor or support visa transfers.

Requirements

  • Experience shipping LLM-based systems that people rely on in production, rather than demos or research-focused fine-tuned models.
  • Ability to identify failure modes in agent systems and design around them.
  • Ability to own work end to end, from rough prototype through production-grade engineering, without relying on a handoff queue.
  • Current, informed judgment about recent changes in agent design and how they should affect implementation choices.
  • Deep understanding of agent architecture, including how a fast, high-cardinality data store affects what an agent can reason about and how it should be designed.
  • Product judgment to identify what the agent layer should do next and make a case for it with a prototype.

Nice To Haves

  • Experience with observability or developer tools.
  • Familiarity with evaluation frameworks, agent tooling, retrieval-augmented generation (RAG), and prompt engineering.

Responsibilities

  • Design and deliver production-grade agents that investigate, reason over, and act on live observability data in Canvas, including taking an agent from prototype to production readiness.
  • Scope, build, ship, and maintain agent capabilities, including evaluation systems that measure whether changes improve agent performance.
  • Contribute to full-stack development in addition to agent-focused work.
  • Use Honeycomb’s high-cardinality data store to enable agents to correlate across services, investigate traces, and compare system behavior before and after deployments.
  • Extend capabilities in Canvas, the MCP server, and Canvas Skills, including areas such as memory, spatial awareness, and improving the Bedrock loop.
  • Identify and propose future product capabilities with working prototypes, and distinguish promising developments from hype.
  • Define standards for agent quality, including measurable evaluation, maintainability, and honest communication of limitations.

Benefits

  • Generous equity with an employee-friendly stock program.
  • Transparent, experience-based pay levels.
  • Unlimited paid time off.
  • Home office, co-working, and internet stipend.
  • Full benefits coverage for employees, with additional coverage available for dependents.
  • Up to 16 weeks of paid parental leave, regardless of path to parenthood.
  • Annual development allowance.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service