Senior AI Implementation Engineer

Overview

At Lindy AI, I work at the layer where demos become products. I build the evals infrastructure that measures agent task success, harden reliability so agents don't fail silently, engineer context management for long-running workflows, and author the skills and routines that thousands of users' agents execute daily.

What I did

• Build and own evals infrastructure that turns agent quality from vibes into measurable gates.
• Design context management systems keeping multi-step agents coherent over long-running tasks.
• Author platform-wide skills and routines — the playbooks thousands of users' agents run on.
• Engineer reliability: guardrails, verification loops, structured outputs, and safe tool execution.
• Work forward-deployed with enterprise customers, scoping and hardening agents against real-world messiness.

Outcomes

• Moved agent quality from anecdote to measurement with evals gating production changes.
• Reduced silent failure modes through tighter schemas, validation, and tool permissions.
• Skills and routines I wrote are executed by thousands of agents across customer workflows.