Nehaveigur

The genetics research team I manage has been experimenting with AI agents for a few months now. Within my biotech company, we’re well suited to drive the implementation of AI since we have both the ability and the motivation. We know how to code and we are good at dealing with large datasets, but we also have to deal with interdisciplinary, unstructured data and incomplete information and are often called upon to provide answers within hours.

The biggest challenge to make the agents do useful work has been to give them access to relevant files and information on our company network. This includes research papers, emails, notebooks, internal databases and thousands of internal PowerPoint decks that have accumulated over the last decade. All the work we have put into archiving our work and making it discoverable is now paying off. Austin Vernon describes the challenge like this:

Human organizations rarely codify their entire structure because the upfront cost and coordination are substantial. The ongoing effort to access and maintain such documentation is also significant. Asking co-workers questions or developing working relationships is usually more efficient and flexible.

Asking humans or developing relationships nullifies AI’s strength (speed) and exposes its greatest weakness (human context). Having the information in written form eliminates these issues. The cost of creating and maintaining these resources should fall with the help of AI.

I‘ve written about how organizations already codify themselves as they automate with traditional software. Creating wikis and other written resources is essentially programming in natural language, which is more accessible and compact.