Hiring: Benjamin RL

n

nWe’re training the next generation of Vedika models.

n

nThis role sits directly inside RL and post-training research: designing how the model learns after pretraining, how it improves from interaction, how it handles long-horizon tasks, and how we push capability beyond standard instruction tuning.

n

nYou’ll work on:

n• RL training for next-generation Vedika models

n• GRPO, PPO, DPO and newer post-training methods

n• Reward models, process rewards and verifiers

n• Long-horizon reasoning and agent trajectories

n• Tool-use and computer-use reinforcement

n• Self-improvement and synthetic training loops

n• Multi-turn behaviour and memory training

n• Failure mining from model trajectories

n• Evaluation systems for reasoning, autonomy and reliability

n• Research experiments that can become part of the next model generation

n

nCompensation:

n₹2.6 LPA fixed

n₹3.6 LPA CTC

n

nWork mode:



Fully remote

n

nYou’ll get:

n• Mac for development

n• Claude

n• Codex

n• Reliable compute and research infrastructure

n• ₹10L–₹50L+ yearly AI/token spend available across the team and experiments

n

nThis is not a role for someone whose idea of model work ends at prompting or basic fine-tuning.

n

nWe want someone who can understand a training run, break it, diagnose it, redesign it and make the next model measurably better.

n

nStrong PyTorch, RL fundamentals, post-training, distributed training and hands-on experimentation matter far more than credentials.

n

nRole: Benjamin RL

nVedika — Next Generation Models

n

nSend your work, experiments, papers, repos or anything you trained that genuinely got better.

📌 Benjamin RLHF (Pune)
🏢 Vedika API
📍 Pune