Research Intern, Agent RL Training

Research Intern, Agent RL TrainingMountain View, California, United StatesAbout NewsBreakFounded in 2015, NewsBreak is the Content Intelligence platform shaping the future content economy. With over 40 million monthly active users, our flagship platform delivers highly personalized local news and information powered by advanced AI, recommendation systems, and adtech.Recognized by Fast Company as #32 on the Top Workplaces for Innovators, we're proud to be Great Place to Work® certified and home to a dynamic team of technologists, product innovators, and business leaders who are passionate about solving meaningful challenges at scale.Together, we reached unicorn status in 2021, and we remain committed to continuing this high-growth trajectory with the right team to fulfill our mission: building the infrastructure layer for content intelligence.If you're inspired to dream big, innovate fast, and make a difference, we'd love to hear from you! For more information, visit the RoleWe are looking for a Research Intern to join our Agent RL Training team. You will be paired with a full-time employee as your mentor, working together to explore, from zero to one, how to apply large language models to NewsBreak's core business, including content understanding, recommendation, agentic web browsing, and autonomous multi-step task completion.This is a hands-on research role. You are expected to independently drive experiments, propose novel ideas, and iterate quickly. We value self-starters with deep intellectual curiosity and the drive to push boundaries in LLM post-training and agent capabilities.Location: Onsite in Mountain View, CA officeWhat You'll Work OnCollaborate with your full-time mentor to identify high-impact research directions for applying LLMs to NewsBreak's productsIndependently run end-to-end SFT experiments on LLM-based agents, and assist with RL-related exploration such as reward design and training iterationCurate and build high-quality training datasets: instruction-following, preference pairs, agent trajectories, and synthetic dataContribute to public publications; we encourage and support top-venue submissions during your internshipWhat We're Looking ForRequirementsHighly motivated and committed: willing to put in extra hours when needed to push projects across the finish lineGenuine passion for research: you read papers for fun, tinker with models on weekends, and care deeply about advancing the fieldIndependently capable of end-to-end model SFT: with basic understanding of RL-based post-training methods (RLHF, DPO, PPO, GRPO, etc.)Excellent taste in model behavior: able to reason about what "good" looks like across user-facing domains and articulate whyStrong Python and PyTorch skillsPreferred QualificationsPublication at a top-tier venue (NeurIPS, ICML, ICLR, ACL, EMNLP, or equivalent)Experience with multi-node distributed training (FSDP, DeepSpeed, Megatron-LM)Proficiency in writing custom GPU kernels with Triton or CUDAExperience building synthetic data pipelines for agent trainingFamiliarity with open-source RL frameworks: TRL, OpenRLHF, veRL/vLLMHourly Pay: $35- $50The US base salary range for this full-time position is listed below. Pay may vary based on a number of factors including job-related skills, level, experience, geographic location and relevant education or training. At NewsBreak, we design our overall rewards package to attract top talents. Depending on the position, the role may also be eligible for discretionary bonus and options. Your recruiter can share more details during the hiring process.Annual Base Pay Range$35 - $50 USDCPRA Privacy Notice for California Candidates