Join CuspAI as an Applied AI/ML Research Engineer to develop generative models for next-generation materials discovery.
Posted by employer 1 day ago
First seen on Joblaze 1 day ago
Last verified on the company career page 1 day ago
What you'll build
Must have
Nice to have
Practical constraints
AI in the day-to-day
We are harnessing AI to unlock trillion-dollar materials breakthroughs.
Not disclosed in this posting: compensation, years of experience, visa sponsorship.
Benefits
Joblaze summary
The Applied AI/ML Research Engineer at CuspAI focuses on developing and evaluating generative models for material discovery, specifically targeting structures that meet defined properties and constraints. Proficiency in building diffusion, flow, or autoregressive models, along with strong software engineering skills in Python and frameworks like PyTorch or JAX, is essential. This role is suited for individuals with a collaborative mindset and a background in AI or materials science, particularly those eager to contribute to sustainability challenges. The position involves close interaction with a diverse team of chemists and engineers, emphasizing interdisciplinary collaboration.
Joblaze insights
Quick facts
From the original posting
We are looking for an Applied AI/ML Research Engineer* to build, train, and evaluate generative models in our materials discovery platform: diffusion, flow, and autoregressive models that generate material structures according to a prompt of target properties and design constraints.
*Note that you would be joining as a ‘Member of Technical Staff’, but the job title above hopefully helps to explain the nature of this role.
You will contribute to all aspects of model development: processing training data, defining evals, coming up with hypotheses, implementing models, building training infrastructure, and making sure that the final models are useful in real-world material discovery campaigns. You may be working with inorganic crystals, molecules, polymers, metal-organic compounds, and other material classes. You will be part of a small dedicated team, but interact with a large group of chemists, machine learning research, and engineers.
In this role you will be building novel generative models that transform how we design new materials, which is critical for accelerating the discovery of next-generation materials for energy and sustainability challenges.
The models you will build will be used to design materials for better solar cells, catalysts, semiconductors, batteries, carbon-capture sorbents, and so on.
Generative Model Development & Evaluation
Develop, train, and rigorously evaluate generative models of material structures across different material classes and modalities.
Define and implement evaluations for material generators that measure what really matters for actual material discovery problems.
Assemble, process, and check material data that may be relevant for future models.
Integration, Engineering & Discovery Campaigns
Integrate your models into the wider CuspAI platform, ensuring they are robust, scalable, and accessible for material discovery workflows.
Collaborate with the software engineering team to adhere to best practices in coding, testing, and deployment.
Support material discovery campaigns that use your generative tools to identify promising candidates for real-world applications, for instance in solar cells, semiconductors, catalysts, or carbon capture.
Analyse the outputs of these campaigns to iteratively improve evals, data, and models.
Interdisciplinary Collaboration
Work together with the existing material generation team and the wider Cusp technical team, to align model capabilities with experimental realities.
Partner with material scientists and chemists to understand the physical constraints and properties required for valid and novel materials generation.
Generative models: You possess deep experience building, training, and evaluating diffusion, flow, or autoregressive models, ideally at scale.
Software engineering: You are a strong software engineer and have experience developing code in a team. You take pride in building robust and maintainable tools ready for real-world use. You know your way around Python and either PyTorch or jax.
Collaborative spirit: You have a demonstrated ability to work well in a team, communicating complex technical concepts to colleagues from different scientific backgrounds.
Mission alignment: A genuine enthusiasm for using technology to do something good in the world and solve sustainability challenges.
We would love you to have experience with at least one of the following:
Multimodal generation: You have built, trained, or evaluated multimodal generative models.
Reinforcement learning: You have experience using reinforcement learning, ideally in the post-training of generative models.
Domain expertise: You have experience with problems in chemistry or material science, for instance by previously building models in this domains. You know your perovskite from your polymer. A formal education or degree on these subjects is not required.
This role could be based in our Amsterdam, London, Cambridge, or Berlin offices, with the expectation of being in the office three days per week. Additionally, there may be regular travel required to other locations for collaboration and project work.
A competitive salary: We value and reward impact and growth
Standard company text repeated across CuspAI's postings is omitted here.