Senior ML Scientist Reinforcement Learning

Job Description

Location Details: Gurugram At GoDaddy the future of work looks different for each team. Some teams work in the office full-time; others have a hybrid arrangement (they work remotely some days and in the office some days) and some work entirely remotely. This is a hybrid position. You?ll divide your time between working remotely from your home and our Gurugram office, so you should live within commuting distance. Hybrid teams work in-office 3 days each a week. The hiring manager can share more about what hybrid work might look like for this team. Join our Team? If you?re passionate about machine learning and want to build systems that learn and adapt over time, join our Data and Artificial Intelligence (DnAI) team at GoDaddy as a Senior Machine Learning Scientist focused on reinforcement learning. This is a new and exciting role crafting intelligent agents that interact with multifaceted environments, learn from feedback, and continuously optimize their decision-making. Working alongside world class engineers and scientists to bridge experimentation, data, and real-time adaptation ? driving smarter, more personalized customer experiences. If you are excited about reinforcement learning, have depth in A/B Testing analysis and ML modeling, we encourage you to apply. What you'll get to do.. . Develop and implement reinforcement learning algorithms that enable agents to learn optimal policies from experimentation outcomes and user interactions Transform experimental data into reward signals and build continuous learning loops that improve long-term business and customer outcomes Develop robust training and evaluation pipelines, including offline simulation, online experimentation, and scalable deployment environments Integrate Reinforcement Learning-based decision systems into production, collaborating with multi-functional partners to embed adaptive learning into products and platforms Develop experiments including offline simulations, online A/B tests, and evaluation pipelines to validate RL/LLM models, ensuring robustness and business value Explore sophisticated methods like contextual bandits, model-based reinforcement learning, DPO for language-driven problems Fine-tune LLMs using RLHF techniques, redefining preference data into reward signals and building iterative training loops to enhance agent performance and alignment Your experience should include... 6+ years in ML focused on reinforcement learning with a deep understanding of LLM and fine-tuning methods, including supervised fine-tuning (SFT), parameter-efficient tuning (LoRA/PEFT), and alignment techniques in a global organization Experience with reinforcement learning for language models (e.g., RLHF with PPO, Direct Preference Optimization, reward modeling from human or AI feedback) Experience in experimentation and evaluation for reinforcement learning enhanced LLMs, including offline simulations, online A/B tests, bandit allocation, and guardrail metrics for safety, bias, and hallucination Strong Python skills and proficiency with ML frameworks (e.g., PyTorch, TensorFlow) and software engineering protocols (version control, CI/CD, code review) Experience defining reward functions and evaluation metrics for reinforcement learning systems in real-world environments You might also have... Master?s or PhD in computer science, engineering, statistics, or a related field Track record of deploying reinforcement learning agents at scale in production Proficiency in Typescript or experience with environment modeling tools