RL Lead @ Deccan AI | Post training, SFT, RLHF, LLM Evals, RL Gym · Deccan AI
1mo agoon tal BOSS
Software Engineer- RL EnvironmentsDeccan AI
Hyderabad · hybrid · 2-4 yrs · 20-40 LPA
Deccan AI develops research-grade evals and benchmarks for frontier AI labs. As a Software Engineer for RL Environments, you will bridge the gap between research and production by building the core infrastructure used to train and evaluate AI agents. You will collaborate with research teams to define task logic, design reward signals, and deploy verifiers that support our agentic and robotics initiatives.
what you'll do:
- Design and build scalable environments for training and evaluating reinforcement learning agents.
- Define task requirements and reward structures in collaboration with research scientists.
- Develop reliable verification systems to measure agent performance across diverse domains.
- Implement agentic workflows and tool-use integration within our simulation platforms.
- Optimize environmental performance to support large-scale model training and benchmarking.
- Maintain clean and robust codebase standards to ensure reproducibility in experiments.
why join: Deccan AI works at the frontier of model performance, partnering with leading AI labs and enterprises. Joining the team allows you to build foundational tooling for AI agents and evaluation at scale, supported by a clear mission to improve model accuracy and reliability.
message me directly on tal BOSS.
tal saysbuild scalable simulation environments and agentic benchmarks to advance reinforcement learning at deccan ai.
engineering
mid
hybrid
Hyderabad
pythongoreinforcement learning
comments 1 anon