---
title: 'Research Scientist, Scaling RL at Periodic Labs'
canonical: 'https://feeny.ai/job/research-scientist-scaling-rl-periodic-labs-menlo-park-g9s9gctkbnjs'
type: 'job'
last_seen: '2026-10-19'
---

# Research Scientist, Scaling RL at Periodic Labs

- **Company:** Periodic Labs
- **Location:** Menlo Park, CA / Montréal, Canada
- **Employment:** full-time
- **Posted:** 2026-10-01
- **Last confirmed live:** 2026-10-19
- **Apply:** https://jobs.ashbyhq.com/periodic-labs/20b122c9-b8ec-4fb0-aaf4-9b45902affe0

## Job description

## About Periodic Labs

We're an AI and physical sciences company building state-of-the-art models to accelerate breakthroughs across materials, energy, and beyond. Backed by world-class investors and growing rapidly, we operate at the pace the frontier requires. Our team brings deep expertise, genuine ownership, and a drive to push the boundaries of what's scientifically possible.

## About the Role

We're training frontier models to develop deep scientific knowledge and reasoning for scientific tasks. You’ll study how RL scales with training compute, develop better algorithms, and take ideas from controlled experiments to our largest runs like [Periodic Neon](https://periodic.com/news/nature-is-our-learning-environment).

## What You'll Do

- Design experiments to understand how RL performance scales with compute, model size, data, and reward quality, building on work such as[ScaleRL](https://arxiv.org/abs/2510.13786)
- Develop better RL algorithms, spanning policy optimization, advantage estimation, exploration, and credit assignment for long-horizon RL tasks
- Build adaptive sampling and curriculum methods that adjust task difficulty, problem selection, and the number of rollouts as models improve
- Study bias and stability during RL training, including importance-sampling corrections and methods to tackle policy staleness and training–inference mismatch, as [discussed here](https://www.youtube.com/watch?v=GH4JCdAAUYg).
- Improve compute efficiency across training and inference through experiments with hyperparameters, such as length penalties, rollout counts, batch sizes, and update schedules.

You Will Thrive in This Role If You Have

- Hands-on experience training LLMs with reinforcement learning
- Strong attention to detail and rigorous approach to answer questions scientifically.
- Coming up with small-scale RL setups that transfers to large-scale training runs.
- Comfort working across a complex training stack to implement, debug, and test new research ideas.
- We are open to offering a 1 year post-doc for this profile. Early career (final year PhD students) are encouraged to apply.

Mechanics Minimum education: Bachelor’s degree or similar experience Location: Menlo Park, CA or Montreal, Canada Compensation: $225,000-$350,000 base + equity Visa sponsorship: Yes, we sponsor visas and will do everything we can to assist in this process.

## About Periodic Labs

## Company Overview
- **One-liner**: Periodic Labs is building AI scientists and autonomous laboratories to accelerate scientific discovery in the physical sciences, starting with materials design and semiconductor research.
- **Entity Type**: Private (seed-stage, $300M total funding raised in 2025)
- **Headquarters**: San Francisco, California, USA (primary) and Menlo Park, California, USA (dual locations) [cbinsights.com](https://www.cbinsights.com/company/periodic-labs), [linkedin.com](https://www.linkedin.com/company/periodic-labs)
- **Founded**: 2025 [cbinsights.com](https://www.cbinsights.com/company/periodic-labs)
- **Founders**: Ekin Dogus Cubuk (Co-Founder) and other undisclosed founding team members with backgrounds at OpenAI, DeepMind, and Google [linkedin.com](https://www.linkedin.com/company/periodic-labs), [periodic.com](https://periodic.com/)

## Core Business
- **Primary industry/industries**: Physical sciences R&D, artificial intelligence, autonomous laboratory systems
- **Target customers**: B2B – enterprises in semiconductor manufacturing, materials science, energy, and other deep-tech sectors requiring experimental research
- **Mission or purpose statement**: “Our goal is to create an AI scientist” that can autonomously hypothesize, run experiments, and learn from results to discover new materials and accelerate technological progress [periodic.com](https://periodic.com/)

## Products & Services
- **AI Scientist**: A suite of AI models trained to generate hypotheses, design experiments, and interpret results across physical science domains.
- **Autonomous Laboratories**: Fully automated, high-throughput experimental platforms that produce large volumes of high-quality data (including negative results) to train AI models. [periodic.com](https://periodic.com/)
- **Custom Agents for Industry**: Tailored AI agents for engineers and researchers, e.g., helping a semiconductor manufacturer analyze heat dissipation data to accelerate iteration. [periodic.com](https://periodic.com/)

## Market Standing
- **Valuation/Market Cap**: Not disclosed (private seed-stage company)
- **Key Metric**: Total funding raised – $300M in a single seed round (2025) [cbinsights.com](https://www.cbinsights.com/company/periodic-labs)
- **Notable Investors/Partners**: a16z, Felicis, DST, NVentures (NVIDIA’s venture arm), Accel, Emerson Collective, Fellows Fund, Coatue; individual backers include Jeff Bezos, Elad Gil, Eric Schmidt, Jeff Dean. Also collaborating with academic advisors from Stanford, Northwestern, and more. [periodic.com](https://periodic.com/), [cbinsights.com](https://www.cbinsights.com/company/periodic-labs)
- **Growth Signals**: 40 employees as of mid-2026, with monthly headcount growth of +19.6% and 25 active job postings (+47.1% quarterly growth) [linkedin.com](https://www.linkedin.com/company/periodic-labs)

## Competitive Advantages
- **Proprietary data generation**: Autonomous labs produce massive, unique datasets (including valuable negative results) that do not exist in public literature, creating a data moat.
- **Founding team pedigree**: Core team co-created ChatGPT, DeepMind’s GNoME, OpenAI’s Operator (Agent), MatterGen, and scaled autonomous physics labs – rare concentration of AI + science talent. [periodic.com](https://periodic.com/)
- **End-to-end integration**: AI models, simulation, and physical experimentation are all in-house, enabling rapid closed-loop learning.

## Strategic Focus
- **Scale autonomous labs**: Expand capacity to generate experimental data across multiple physical science domains.
- **Target high-impact problems**: Superconductor discovery, semiconductor heat dissipation, and other materials challenges that can unlock Moore’s Law, advanced energy grids, and space travel. [periodic.com](https://periodic.com/)
- **Develop the first generation of AI scientists**: Train models that can operate independently of human input across the entire scientific method.

## Why Work Here
- **Culture**: “From bits to atoms” – a deep-tech startup blending frontier AI with hands-on lab work. Emphasis on rapid iteration and real-world impact.
- **Work policy**: In-office preferred; primary offices in Menlo Park and San Francisco, CA. Most roles are on-site. [builtin.com](https://builtin.com/company/periodic-labs)
- **Notable perks**: Work alongside world-class AI researchers and experimental scientists; access to cutting-edge automated lab equipment; early-stage equity and high-growth trajectory.
- **Engineering culture**: 34% of employees are in technical roles; talent sourced from OpenAI, Google DeepMind, Meta, xAI, and Stanford – strong peer learning environment. [linkedin.com](https://www.linkedin.com/company/periodic-labs)

## Sources
1. [periodic.com](https://periodic.com/)
2. [builtin.com](https://builtin.com/company/periodic-labs)
3. [cbinsights.com](https://www.cbinsights.com/company/periodic-labs)
4. [linkedin.com](https://www.linkedin.com/company/periodic-labs)
5. [jobs.ashbyhq.com/periodic-labs](https://jobs.ashbyhq.com/periodic-labs)

## Other roles at Periodic Labs

- [Head of Environmental Health & Safety (EHS)](https://feeny.ai/job/head-of-environmental-health-safety-ehs-periodic-labs-menlo-park-k9rf9q4mgdth) — Menlo Park, CA
- [Process Safety Engineer](https://feeny.ai/job/process-safety-engineer-periodic-labs-menlo-park-kzcpxh3d9mg6) — Menlo Park, CA
- [Controller](https://feeny.ai/job/controller-periodic-labs-menlo-park-y8xmsezeq3ch) — Menlo Park, CA
- [Automation Engineer](https://feeny.ai/job/automation-engineer-periodic-labs-menlo-park-e7a6p1rtctm5) — Menlo Park, CA
- [Research Scientist/Research Engineer, Midtraining](https://feeny.ai/job/research-scientist-research-engineer-midtraining-periodic-labs-menlo-park-gh5qfkb8vxth) — Menlo Park, CA
- [XRD Facility Manager](https://feeny.ai/job/xrd-facility-manager-periodic-labs-menlo-park-artzbwnnmkjx) — Menlo Park, CA
- [Equipment Engineer](https://feeny.ai/job/equipment-engineer-periodic-labs-menlo-park-27e2ekxdqtbm) — Menlo Park, CA
- [Don't See Your Role? Apply Here!](https://feeny.ai/job/don-t-see-your-role-apply-here-periodic-labs-menlo-park-btzy1epyrqbp) — Menlo Park, CA
- [Head of People](https://feeny.ai/job/head-of-people-periodic-labs-menlo-park-5n8dyr2jhzga) — Menlo Park, CA
- [Computational Scientist, Differentiable Physics](https://feeny.ai/job/computational-scientist-differentiable-physics-periodic-labs-menlo-park-0y6vkjnce8d0) — Menlo Park, CA