Research Intern, Agent RL Training

Job description

<div class="content-intro"><p><strong>About NewsBreak</strong></p>
<p><strong>Local life runs on local information.</strong> What happened down the street? Who can fix this? Where should we go tonight? NewsBreak started in 2015 to answer these questions, and today more than 40 million people a month rely on us to follow what’s going on in their zip code.</p>
<p><strong>Three platforms with one vision: powering local life.</strong> Led by founder and CEO Jeff Zheng, NewsBreak is our local life and news platform; Nearby AI connects people with local businesses; and NDot AI helps publishers and creators grow and earn.</p>
<p><strong>Building for the long term.</strong> A $115 million Series C made us a unicorn in 2021, and we keep investing in our people. We’re Great Place to Work® certified and named to <em>Fast Company</em>’s Best Workplaces for Innovators (2025) and <em>Fortune</em>’s Best Workplaces in the Bay Area™ (2026).</p>
<p><strong>Everyone makes an impact.</strong> Whether you write code, shape products, tell our story, or run the business, we want people who own projects end to end, stay curious, and speak up. Join us.</p>
<p>Learn more at: NewsBreak, Nearby AI, NDot AI</p></div><h2>About the Role</h2>
<p>We are looking for a <strong>Research Intern</strong> 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.</p>
<p>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.</p>
<p><strong>Location: </strong>Onsite in Mountain View, CA office</p>
<h2>What You’ll Work On</h2>
<ul>
<li>Collaborate with your full-time mentor to identify high-impact research directions for applying LLMs to NewsBreak’s products</li>
<li>Independently run end-to-end SFT experiments on LLM-based agents, and assist with RL-related exploration such as reward design and training iteration</li>
<li>Curate and build high-quality training datasets: instruction-following, preference pairs, agent trajectories, and synthetic data</li>
<li>Contribute to public publications; we encourage and support top-venue submissions during your internship</li>
</ul>
<h2>What We’re Looking For</h2>
<h3><strong>Requirements</strong></h3>
<ul>
<li><strong>Highly motivated and committed: </strong>willing to put in extra hours when needed to push projects across the finish line</li>
<li><strong>Genuine passion for research: </strong>you read papers for fun, tinker with models on weekends, and care deeply about advancing the field</li>
<li><strong>Independently capable of end-to-end model SFT: </strong>with basic understanding of RL-based post-training methods (RLHF, DPO, PPO, GRPO, etc.)</li>
<li><strong>Excellent taste in model behavior: </strong>able to reason about what “good” looks like across user-facing domains and articulate why</li>
<li>Strong Python and PyTorch skills</li>
</ul>
<h3><strong>Preferred Qualifications</strong></h3>
<ul>
<li>Publication at a top-tier venue (NeurIPS, ICML, ICLR, ACL, EMNLP, or equivalent)</li>
<li>Experience with multi-node distributed training (FSDP, DeepSpeed, Megatron-LM)</li>
<li>Proficiency in writing custom GPU kernels with Triton or CUDA</li>
<li>Experience building synthetic data pipelines for agent training</li>
<li>Familiarity with open-source RL frameworks: TRL, OpenRLHF, veRL/vLLM</li>
</ul>
<p>Hourly Pay:  $35- $50 </p><div class="content-pay-transparency"><div class="pay-input"><div class="description">The 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.</div><div class="title">Annual Base Pay Range</div><div class="pay-range"><span>$35</span><span class="divider">—</span><span>$50 USD</span></div></div></div><div class="content-conclusion"><p><a href="https://docs.google.com/document/d/15apsJLgHq1p5NJR3fzFmdctKX4QIzzf2/edit?usp=sharing&ouid=110601223912453011659&rtpof=true&sd=true" target="_blank">CPRA Privacy Notice for California Candidates</a></p>
<p> </p></div>