Jobs at HeyMilo AI

Manage US payroll operations for a distributed team: run pay cycles, maintain and audit employee records, ensure multi-state compliance, reconcile runs, document processes, and define quality standards. Work closely with operations and finance to prevent payroll errors and handle filings and year-end reporting.

Own Canadian payroll operations: run regular and off-cycle payroll in OrangeHRM, maintain and audit employee records, ensure CRA and provincial compliance, reconcile pay runs, produce ROEs and year-end T4/T4A reporting, document processes, set quality standards, and review payroll accuracy.

Design and build reproducible reinforcement-learning environments, simulators, reward functions, and evaluation harnesses for HR/recruiting workflows. Run evaluations and post-training experiments to validate learnable signals, package results for reproducibility, and collaborate on research write-ups.

Design realistic multi-step evaluation tasks and rubrics for LLMs, build and curate ground-truth datasets, review and analyze model failures, calibrate task difficulty, report score distributions, and contribute to research write-ups.

Design and build reproducible RL environments, verifiable reward functions, and evaluation harnesses for LLMs. Run post-training experiments, package reproducible results, and contribute to research and published evaluations while helping pick new use cases.