Michael
@jaco
Michael
@jaco
What simple experiments or lessons from your research should they start applying right now — before agentic systems become mainstream?
Start by running controlled experiments in simulated environments, such as Magnetic Marketplace, before deploying agents in real-world settings. Focus on stress-testing areas where current models struggle: handling choice overload, negotiating effectively, and collaborating toward shared goals. Use diagnostic tools such as the MCP Interviewer to identify tool interference and protocol issues early. Just as important, design with transparency and human oversight in mind, building layers that allow users to monitor and intervene. But most importantly, developing AI responsibly starts by asking questions about what can go wrong with the systems we are developing, how to test them for such risks, and how to mitigate them. By asking these questions throughout the lifecycle of the systems we develop, we can uncover rough edges and inefficiencies—and ensure that as agentic systems scale, they do so with reliability, fairness, and trust at their core.
Plastic is an engineering-driven AI lab building at the intersection of machine learning and cognitive science.
Our focus is developing Honcho, an AI-native memory solution powered by our state-of-the-art reasoning models. Honcho is a continual learning system for modeling personal identity, and soon a shared context layer for individual alignment.
The foundational layer of intelligence being built is just the beginning. Latent among the scores of specialized secondary and tertiary layers yet to be realized exists one for personal identity.
Stagehand is a browser automation framework used to control web browsers with natural language and code. By combining the power of AI with the precision of code, Stagehand makes web automation flexible, maintainable, and actually reliable.
A members-only community for local media operators to connect, learn, and level up… If you ain't serious, don't join
Local newspapers are dying and algorithms have failed. There's never been a better time to launch your local media company — and build a profitable business, on your terms.