Research Engineer

Founding & Early Research Roles | San Francisco / Bay Area (On-site) | $160K–$450K + Meaningful EquityThe frontier labs get the headlines. The small, elite teams building what those labs depend on get the hardest problems, the most ownership and the equity to match.I'm hiring for several seed and Series A AI research companies, all based in the Bay Area, all backed by top-tier investors, and all with teams of fewer than 15 people. Here's what they're building:RL environments and benchmarks that frontier labs use to train and evaluate computer-use agentsPost-training research to prove which data actually makes models better, through controlled SFT and RL experimentsAI security, post-training models with RL for adversarial simulation and red-teamingClinical AI that reads medical records and makes defensible decisions on healthcare paymentsLong-horizon agent simulation, generating the worlds and verifiers that decide whether an agent really succeededYou might be a fit if you:Have hands-on experience in post-training (SFT, RLHF, RL), evals, reward modeling, agentic systems or computer-use modelsCome from a rigorous research background, such as a leading CS/ML PhD or MS program, a top AI lab, or published work at venues like NeurIPS, ICML or ICLRHave shipped models or research systems into production, not just papersWould rather own a problem end to end than be one of 200 people on a model cardWhat's on offer:$160K–$450K base depending on role and seniority, plus 0.2%–1.5% equityFounding or early-team seats working directly with founders from DeepMind, Meta, Stanford and top quant firmsOn-site in SF or the Bay Area, with visa sponsorship or transfers available for several rolesWhether you're finishing a PhD, leaving a postdoc, or ready to trade a big-tech title for real ownership, I'd love to talk.#AppliedAI #AIResearch #PostTraining #ReinforcementLearning #LLM #Evals #AIAgents #MachineLearning #PhD #Stanford #MIT #CMU #Berkeley #Hiring #DigitalWaffle