The company's tool offers data enrichment, fine-tuning, explainability, and the generation of new sequences for improved outcomes. Converge Bio uses open-source foundation models and is also developing its own. The company plans to use the funding to hire more staff, acquire customers, publish a scientific paper on antibody design, and train a proper foundation model. The seed round was led by TLV partners.
Key takeaways:
- Converge Bio has developed a tool to help companies implement biology-focused large language models (LLMs) in their research, and has raised $5.5 million in a seed round to scale its product.
- The company's tool helps to enrich data, explain answers, and make LLMs more usable in specialized domains like microbiology or immunology.
- Converge Bio uses open source and free foundation models, but is also working on creating its own. It has a proprietary process for the explainability part of its tool and a unique data enrichment "curriculum".
- The company plans to use the funding to hire more staff, acquire customers, publish a scientific paper on antibody design, and train a proper foundation model.