Senior Software Engineer II - Agentic Intelligence

Honeycomb is a service for the near and present future, defining observability and raising expectations of what developer tools can do. We work with well-known companies like HelloFresh, Slack, LaunchDarkly, and Vanguard across a range of industries. This is an exciting time in our trajectory—we’ve closed Series D funding, scaled past the 200-person mark, and were named to Forbes’ America’s Best Startups of 2022 and 2023.

About the Role

AI agents at Honeycomb investigate, reason, and act on real observability data. They live in Canvas: the agentic workspace where engineers go to understand their systems. The Agentic Intelligence team has shipped Canvas, the Honeycomb MCP server, and the Canvas Agent and Canvas Skills surfaces.

We're looking for someone who brings deep agent expertise to expand what the team can build: new agents, new surface area in Canvas, memory, spatial awareness, and improved performance on our Bedrock loop. Honeycomb's data store is fast and accepts high-cardinality data, enabling agents built on it to do things no other observability product can. This role is about leveraging that capability to build agents that push the boundaries of what’s possible.

Some of this work will start as a prototype, with the expectation that the code evolves into something production-ready.

Responsibilities

• Design and deliver production-grade agents that investigate, reason, and act on live observability data inside Canvas—trustworthy even in high-pressure situations like mid-incident.
• Own the full lifecycle of agent development: scope, build, ship, and maintain agents, including evals to measure improvement.
• Build agents only Honeycomb can build, leveraging a data store that returns high-cardinality queries in seconds to reason over signals conventional backends can’t serve at this fidelity (e.g., correlating across services, drilling into traces, comparing pre- and post-deploy states).
• Extend the surface area of Canvas, the MCP server, and Canvas Skills (e.g., memory, spatial awareness, a faster Bedrock loop) and advocate for what’s next with working code.
• Define what "good" means for agents at Honeycomb: measurable against real evals, maintainable, and honest about their limits.
• Contribute to full-stack development alongside agent-focused work.

Example Projects

• Multiple agents collaborating on a shared Canvas investigation, each claiming a hypothesis, publishing findings, and narrowing the search space for faster resolution (blog).
• Auto-investigation triggered by an SLO burn alert, where the agent forms hypotheses and prepares visualizations before a human looks, cutting mean-time-to-insight for on-call (o11ycon 2026).
• Skills that encode a team’s domain expertise (e.g., Kubernetes thresholds) so agents and human colleagues can leverage them (o11ycon 2026).

Requirements

• AI and agent engineering experience: You’ve shipped LLM-based systems people relied on in production—not demos or research prototypes. You know where agent systems break and how to design around those failures.
• End-to-end ownership: On a small team, there’s no handoff queue. You can take a rough prototype to production-grade without needing someone behind you to do the durable engineering.
• Current judgment: You have informed opinions about how agent design has shifted in the last 6–12 months and how that would change your approach today.
• Agent architecture depth: You understand how a fast, high-cardinality data store changes what an agent can reason about and how to design for it.
• Product judgment: You can assess the agent layer’s current capabilities, identify what it should do next, and make that case with a prototype—not just a deck.

Nice to Have

• Observability or developer-tools background. Engineers are your users; fluency in this world helps you ramp faster and build better.
• Familiarity with eval frameworks, agent tooling, RAG, and prompt engineering.

Benefits

• A stake in our success: generous equity with an employee-friendly stock program.
• Transparent compensation: pay based on levels relative to experience (no negotiation required).
• Unlimited PTO to recharge.
• A distributed-first mindset and culture.
• 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.

Pay

Base salary range: $183,340 – $206,000 USD (based on level of experience).