Foundation and generative models for biomolecules at Inceptive

Berlin, California, Germany -

Full Time

Start Date

Immediate

Expiry Date

27 Jan, 26

Salary

240000.0

Posted On

29 Oct, 25

Experience

5 year(s) or above

Remote Job

Yes

Telecommute

Yes

Sponsor Visa

Skills

Machine Learning, Python, PyTorch, Generative Models, Data Processing, Biochemistry, Molecular Biology, Software Engineering, Model Training, Bioinformatics, Experimental Biology, Data Visualization, Collaboration, Mentorship, Model Evaluation, Compute Cluster Orchestration

Industry

Biotechnology Research

Description

At Inceptive, you will drive forward development that could help billions of people. To accomplish this, you will be part of a collaborative, antedisciplinary team building our biological software. The design space of biomolecules is unimaginably vast — far beyond what can be explored experimentally. Yet within this space lie molecules with properties essential for new medicines. Our machine learning models learn to design therapeutic biomolecules with specific, desirable functions. We advance the state of the art in molecular design by training large-scale foundation models and developing cutting edge generative approaches. The models learn from diverse heterogeneous datasets and are refined through focused fine-tuning and feedback from experiments. Key to progress is a team that combines exceptional machine learning expertise with thorough domain understanding. You will collaborate closely with other machine learning researchers and engineers, as well as computational and experimental biologists, to advance these models and translate their capabilities into real therapeutic designs. Your Mission, should you choose to accept it Embody our vision of an antedisciplinary environment and embrace learning about areas outside of your traditional area of expertise Develop, implement, train, and iteratively improve state-of-the-art models for biomolecule design Analyze, visualize, and communicate results to support team efforts in improving models and data Create, deploy, and refine tools for efficient, reliable machine learning experimentation and production Work with biologists to collect data for the training and evaluation of generative models of biomolecules Provide mentorship and technical direction to team members as appropriate Qualifications 3+ years of hands-on experience developing ML models Demonstrated track record of implementing, training, improving advanced machine learning models Highly capable programmer fluent in Python ecosystem and PyTorch or similar deep learning framework Availability to work with team members across US and Europe, with meetings starting at 8am PT and ending at 7pm CET Readiness to travel several times a year for company retreats and business events We value the benefits of in-person collaboration and expect candidates to primarily work from our office locations Preferred technical skills PhD in AI/ML, computer science, computational biology, physics, or a related field, or equivalent practical experience (e.g., industry experience, research, or advanced technical expertise) Strong skills in designing, executing, and documenting machine learning experiments Practical experience with modern generative models Strong software engineering skills, in particular for data processing, evaluation of ML models, compute cluster orchestration Experience with large-scale model training, foundation models, model parallelism, multi-node training Experience with bio sequence data and datasets — various genomic and protein data, sequencing, functional assays, etc Knowledge of biochemistry, molecular/cell biology, and drug development Compensation $135K – $240K + Bonus + Equity What we offer A competitive compensation package 30 days paid vacation per year Comprehensive health insurance and 401K with company match for US based Beginners, with comparable benefits for those based in other countries Quarterly company-wide retreats Monthly wellness benefit Budget for multiple visits per year to our offices in Berlin, Palo Alto or Switzerland Learning & Development budget to attend conferences, take courses, or otherwise invest in your professional growth, as well as access to the Learning & Development platform EdX and Hone A buddy to help you get settled *Varies by country At Inceptive, we are creating tools to develop increasingly powerful biological software for the rational design of novel, broadly accessible medicines and biotechnologies previously out of reach. Our team brings together vast expertise in molecular biology, machine learning, and software engineering, and we are all working towards becoming antedisciplinary, meaning we deepen the knowledge we have in our area of expertise while also expanding our knowledge of completely new fields. We approach our goals with a Beginner's mind, humbly and with fresh eyes, and aim to become the pioneers of a new discipline rooted in biology as much as in deep learning, whose impact will be realized together with out-of-the-box thinkers in business and entrepreneurship, defying established categorizations. We are building a company culture centered around growth, learning, and discovery. We believe in humility and open-mindedness in how we approach each other, as well as problems we don't yet have solutions for. It is the policy of Inceptive to ensure equal employment opportunity without discrimination or harassment on the basis of race, color, religion, sex, sexual orientation, gender identity or expression, age, disability, marital status, citizenship, national origin, genetic information, or any other characteristic protected by law. Inceptive prohibits any such discrimination or harassment. Inceptive is also committed to welcoming and providing accommodations to people with disabilities. Please let us know if you need any accommodations throughout your application process.

Responsibilities

You will develop, implement, train, and iteratively improve state-of-the-art models for biomolecule design. Collaborating closely with other researchers and engineers, you will analyze and communicate results to enhance model performance.