Our mission is to make biology easier to engineer. Ginkgo is constructing, editing, and redesigning the living world in order to answer the globe’s growing challenges in health, energy, food, materials, and more. Our bioengineers make use of an in-house automated foundry for designing and building new organisms.
AI Enablement is a small team with an ambitious goal: making it as easy as possible to use AI and ML to make biology easier to engineer. We work with biologists, software and devops engineers, and data scientists to create the necessary MLOps infrastructure, organize data, evaluate, train, and fine-tune new models and approaches, consult and prototype – whatever needs doing!
As an AI Enablement Co-op, you will work with software engineers and scientists to train and evaluate new LLMs for biology. You will have the opportunity to work on applying both open-source and internal foundation models to specific problems in bioengineering. You will also work on tools surrounding these workflows, making it easier to create, deploy, and manage both models and data.
Possible Projects Include:
- Deploy a recently published DNA LLM into the cloud environment, and evaluate its performance on a classification task using Ginkgo data.
- Design training and validation splits, clean and prepare dataset, etc.
- Experiment with using natural language LLMs to create internal agents to assist scientists in research, writing, and other suitable tasks.
The Ginkgo Bioworks Early Talent program is open to students who will return to their degree programs upon completion of their employment at Ginkgo. Candidates must be enrolled in a United States based institution and/or have work authorization in the United States to be eligible to participate.
- Coding review
- Compare results of fine-tuning vs in-context learning. Probe the model to analyze what it gets right & wrong.
- Perform ablation experiments to analyze contribution of certain datasets, model architecture features, and other aspects of Ginkgo’s bio-LLMs.
- Propose and test results of model improvements based on latest research.
- Automate benchmarking processes useful for evaluating various types of biological LLMs. Integrate with our training and evaluation infrastructure
- Currently enrolled in a bachelor’s degree program (current junior at time of application), or enrolled in master’s degree or PhD program, with a major or academic concentration in computer science or computational biology, or a related field
- Strong programming skills, particularly in Python
- Evidence of ability to execute in a self-directed manner
- Strong written communication skills
- Curiosity about bioengineering and applying Machine Learning to biology
- Experience with PyTorch
- MLFlow Bioinformatics experience
- Experience with cloud environments like AWS and GCP
We also feel that it’s important to point out the obvious here – there’s a serious lack of diversity in our industry, and that needs to change. Our goal is to help drive that change. Ginkgo is deeply committed to diversity, equity, and inclusion in all of its practices, especially when it comes to growing our team. Our culture promotes inclusion and embraces how rewarding it is to work with people from all walks of life.
We’re developing a powerful biological engineering platform, so we must remain mindful of the many ways our technology can – and will – impact people around the world. We care about how our platform is used, and having a diverse team to build it gives us the best chance that it’s something we’ll be proud of as it continues to grow. Therefore, it’s critical that we incorporate the diverse voices and visions of all those who play a role in the future of biology.
It is the policy of Ginkgo Bioworks to provide equal employment opportunities to all employees and employment applicants.