LLM Researcher

Fully remote with a high budgetAbout the RoleWe are looking for a highly motivated Research Scientist with strong expertise in Large Language Models (LLMs) and Reinforcement Learning (RL). The candidate will drive cutting-edge research at the intersection of reasoning, decision-making, and scalable AI systems, with opportunities to publish at top-tier conferences (NeurIPS, ICML, ICLR, ACL).This role combines fundamental research with applied innovation, bridging advanced model design with practical AI agent capabilities for gaming, automation, and reasoning tasks.⸻Key Responsibilities• Conduct research on LLM reasoning (causal, logical, multi-step inference) and improve state-of-the-art performance.• Design and implement RL-based methods (RLHF, Deep RL, Bandits, policy optimization) to enhance LLM adaptability and decision-making.• Develop and benchmark novel frameworks for AI agents integrating function calling, external tool use, and multi-modal reasoning.• Build evaluation datasets/benchmarks (reasoning, causal inference, alignment) and publish results at major ML/NLP venues.• Collaborate with cross-functional engineering teams to translate research outcomes into production-ready systems.• Stay at the forefront of LLM + RL research trends, guiding long-term technical strategy.⸻QualificationsMinimum Requirements:• MSc/PhD (or equivalent research experience) in Computer Science, Machine Learning, NLP, or related fields.• Strong background in LLMs (GPT, LLaMA, Claude, etc.) and fine-tuning methods (SFT, RLHF, PPO, DPO).• Solid understanding of reinforcement learning (Q-learning, policy gradient, multi-armed bandits, model-based RL).• Proficiency in Python and deep learning frameworks (PyTorch, JAX, TensorFlow).• Research track record: publications in ACL/ICML/ICLR/NeurIPS or equivalent.Preferred:• Experience building evaluation benchmarks for LLMs (reasoning, causal inference, safety, alignment).• Familiarity with LangChain, function calling, AI agent frameworks.• Hands-on experience with distributed training and inference optimization.• Previous experience in industry labs (FAANG, top AI labs) is a plus.⸻What We Offer• Opportunity to work on frontier LLM + RL research with direct applications in game AI, intelligent agents, and reasoning systems.• Competitive compensation (base + performance bonus + research grants for publications).• Publication support at top-tier conferences.