Job description

ABOUT US Edison Scientific builds and deploys AI scientist agents to accelerate science and the development of new medicines. We are an ambitious team run by scientists and engineers from leading institutions across biology, physics, chemistry, and AI. ABOUT THE ROLE As a Member of Technical Staff, AI Research Scientist at Edison Scientific, you play a central role in developing the models and algorithms that accelerate scientific research. Your role will encompass both cutting-edge research and applied methods, bringing your work to bear on scientific challenges in laboratories and scientists around the world. KEY RESPONSIBILITIES - Interpret challenges in building AI agents for science as well-formulated statistical problems - Conduct research in post-training of LLMs on challenging scientific reasoning tasks, with the goal of exceeding expert human capacity - Conduct research in optimal sampling techniques, considering aspects like efficient exploration and quality-diversity - Own research projects end-to-end, from working with scientists to scope the problem all the way to working with engineers to deploy the outcomes - Develop and extend our experimentation platform for internal tools and projects. - Collaborate closely with a multidisciplinary team of AI researchers, engineers, biologists, and chemists, fostering an environment of innovation and discovery. REQUIRED QUALIFICATIONS - 5-10+ years of strong track record of ML research on real-world problems in (but not limited to) reinforcement learning, representation learning, and Bayesian methods - Fluency in PyTorch, Jax, or equivalent framework - Experience with large-scale training and experimentation technologies (e.g. Ray, verl, Megatron-LM, etc.) - Experience working with data and ML systems: EDA, feature engineering, model development, model deployment, and validation in production systems. - Familiarity with modeling in PyTorch, Jax or equivalent framework. - Demonstrated experience with experimentation in academic or industry settings. - Strong programming expertise with the capability to adapt to various technical challenges in the data, ML, and LLM software stack. PREFERRED QUALIFICATIONS - PhD in Machine Learning, Computer Science, or other quantitative field - Familiarity with leveraging and managing distributed systems SALARY $210,000 - $350,000 • Offers equity WHY JOIN US? We're a fast-moving, mission-driven culture where smart people do their best work and actually enjoy doing it. We also offer full-time employees the following: - Competitive salary and equity - Full healthcare coverage; we pay 100% of premiums for you and your dependents - Support for growing families, including a yearly new parent stipend and fertility coverage through Carrot - Mental health support through Rula, our in-network therapist and psychiatrist network with fast availability - 12 weeks of paid parental leave for maternity, paternity, and adoption - Pet care support with a yearly employer-funded stipend for your animal companions - Commuter benefits so you can pay for transit and parking with pre-tax dollars - 401(k) company matching - $300 health and wellness benefit quarterly - Lunch is on us every day you're in the office, and dinner is on us when you're working late - Regular team off-sites and company events Edison Scientific is an equal opportunity employer. We do not discriminate on the basis of race, color, religion, sex, sexual orientation, gender identity, national origin, age, disability, veteran status, or any other protected characteristic under applicable law.

Requirements

Must-haves

Interpret challenges in building AI agents for science as well-formulated statistical problems
Conduct research in post-training of LLMs on challenging scientific reasoning tasks, with the goal of exceeding expert human capacity
Conduct research in optimal sampling techniques, considering aspects like efficient exploration and quality-diversity
Own research projects end-to-end, from working with scientists to scope the problem all the way to working with engineers to deploy the outcomes
Develop and extend our experimentation platform for internal tools and projects.
Collaborate closely with a multidisciplinary team of AI researchers, engineers, biologists, and chemists, fostering an environment of innovation and discovery.
5-10+ years of strong track record of ML research on real-world problems in (but not limited to) reinforcement learning, representation learning, and Bayesian methods
Fluency in PyTorch, Jax, or equivalent framework
Experience with large-scale training and experimentation technologies (e.g. Ray, verl, Megatron-LM, etc.)
Experience working with data and ML systems: EDA, feature engineering, model development, model deployment, and validation in production systems.
Familiarity with modeling in PyTorch, Jax or equivalent framework.
Demonstrated experience with experimentation in academic or industry settings.
Strong programming expertise with the capability to adapt to various technical challenges in the data, ML, and LLM software stack.

Nice-to-haves

PhD in Machine Learning, Computer Science, or other quantitative field
Familiarity with leveraging and managing distributed systems

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

aimlllmmachine learning

Platform: ashby
Effort: quick

Benefits

Benefits detected in the listing.

401(K)

Parental Leave

Wellness