Looking for a job

Ali Hasan Khan

@alihasankhan1

Message

Senior AI/ML Engineer at ProArch who deployed RL policies that improved long-horizon task success by 34%.

United States

Message

At ProArch, I architected a reinforcement learning platform for sequential decision-making, training actor-critic agents that improved long-horizon task success by 34% over the legacy heuristic baseline. I also applied RLHF and DPO to fine-tune large language models, improving human-preference win rate by 27% and reducing hallucination rate by 31%.

Previously, at Synack, I built reinforcement learning agents for adaptive security testing that increased unique vulnerability discovery by 28% per engagement. At Meta, I built Python and SQL analytics pipelines and developed early machine learning models for engagement and churn forecasting.

Experience

Work history, roles, and key accomplishments

Senior AI/ML Engineer

ProArch

Nov 2023 - Jul 2026 (2 years 8 months)

Architected an end-to-end reinforcement learning platform for sequential decision-making, training actor-critic agents with PPO, SAC, and TD3 that improved long-horizon task success by 34%. Applied RLHF and DPO to fine-tune large language models, improving human-preference win rate by 27% and reducing hallucination rate by 31%.

Machine Learning Engineer

Synack

Jun 2019 - Oct 2023 (4 years 4 months)

Built deep reinforcement learning agents for adaptive security testing, training DQN and PPO policies for sequential attack-path selection that increased unique vulnerability discovery by 28% per engagement. Implemented adversarial evaluation suites and red-team versus blue-team scenarios to harden models against evasion and distribution shift.