Staff Agentic Engineer

Please Note: This is a remote position available in the state listed on this job. Additionally, employment with BambooHR is contingent on passing both a background and credit check.

AI at BambooHR: At BambooHR, we believe in leveraging cutting-edge technology to empower people and transform HR. We’re actively integrating AI into our solutions and workflows to enhance efficiency and drive innovation. To that end, we’re looking to our existing team members and future hires to share this forward-thinking mindset: individuals who are curious about AI’s potential, eager to learn and adapt, and ready to explore how intelligent tools can elevate their work along with BambooHR’s impact on setting people free to do great work. Join us in reimagining the future of HR!

Essential Job Duties

BambooHR is on a mission to set people free to do great work. Our software powers the HR work that small and mid-sized businesses depend on every day. We’re looking for software engineers who use AI to build production-quality software. As a Staff Agentic Software Engineer, you own the technical direction of a product area and the patterns multiple teams build on. You define how engineers across teams specify, delegate, and verify agentic work, and you set the standards that keep that work safe at scale.

AI tools and capabilities change rapidly. To stay resilient through change, you prioritize curiosity and continual learning, and you actively apply new best practices. You are accountable for the technical direction of your product area: the reusable patterns teams adopt, the verification practice that keeps agent-produced work safe, and the leverage that direction creates across teams.

What You’ll Do

Your work runs through four loops, each building on the last: specifying intent, delegating execution to agents, verifying and integrating their output, and codifying best practices and processes to compound value.

• Own the technical direction of a product area, setting how multiple teams specify, delegate, and verify agentic work.
• Define the standard for specification quality across teams: what a complete specification contains and what review gates enforce it.
• Establish the framework for when work warrants agentic execution and when it needs direct implementation, building heuristics that raise the floor for every team.
• Drive large-scale efforts by coordinating human and agentic contributors across multiple teams to deliver more than any one team could.
• Set the verification practice that keeps agent-produced work safe at scale, calibrated to risk and blast radius.
• Set the security architecture for agentic work across the product area, partnering with security to implement it: the data-flow boundaries, access controls, and threat models agents operate within.
• Establish the readiness criteria a product area must meet before agentic work is safe to delegate: the context, guardrails, and observability that must be in place.
• Build the scaffolding that teams delegate through: prompts, agent instructions, repo conventions, test harnesses, and playbooks.
• Resolve the highest-risk technical decisions across the product area, where the stakes or the blast radius are greatest.
• Partner with product leadership to shape what gets built across the product area, bringing engineering judgment into roadmap direction.
• Raise the agentic craft of engineers across teams, mentoring on how they specify, delegate, and verify.

What You’ll Need to get the Job done

You will need depth in these areas to set technical direction, standardize practice across teams, and raise the engineers who deliver.

• 8+ years of software engineering experience, with a record of building and shipping customer-facing software in production and ownership of technical direction across a product area.
• Track record of setting agent patterns and AI-powered solutions that multiple teams reuse.
• Depth in designing systems, data models, and architecture that teams build on, including performance and data-flow tradeoffs.
• Demonstrated ability to define specification standards across teams: what complete looks like and how review gates enforce it.
• Command of context engineering: setting how repo and task context is structured so agents execute reliably across teams.
• Proven ability to design agent workflows for parallel execution across teams, with human checkpoints throughout.
• Confidence reading unfamiliar code and running root-cause analys…