Own end-to-end architecture and execution of AI agent builds for enterprise customers, from systems design and initial scoping through implementation, evaluation, and production deployment
Design and engineer the core agentic logic governing agent behavior, applying engineering principles to optimize quality, reliability, and correctness across non-deterministic model outputs
Engineer and validate layered guardrails and supervisory controls to ensure safe, compliant, and predictable agent performance across real-world scenarios
Architect, build, and test integrations with customer systems (e.g., data pipelines, CRMs, ticketing systems), including building the tools, APIs, and workflows needed for reliable deployments at scale
Interface with senior technical stakeholders at customers to define success criteria and system requirements, and drive technical delivery against timelines
Diagnose, debug, and resolve both probabilistic and technical failures through root-cause analysis of execution traces, error logs, and model behavior
Design evaluation and regression-testing frameworks to validate agent behavior against ground truth and guard against performance drift
Translate customer needs into clear internal documentation and run tight feedback loops with Engineering to drive platform improvements
Partner closely with APMs, Engineering, Design, and Go-To-Market teams to deliver consistent, repeatable, best-in-class agent builds