Dipole
Autonomous AI research agents that learn, explore and iterate toward new discoveries.
The founders
AB
The company
Dipole is building autonomous AI research agents — 'co-scientists' designed to learn, explore, experiment and iterate toward new scientific discoveries. The aim is to automate and accelerate discovery, experimentation and analysis across research fields.
AI agents are moving from chat to doing, and scientific R&D — slow, expensive and iteration-heavy — is an obvious place to point them. Dipole is betting the next frontier isn't answering questions, but running the experiments.
Founded in 2026 and based in San Francisco, Dipole is the work of engineer Aman Bhardwaj, whose background spans software development at Amazon Web Services, Barclays and Cisco. The company is still early — its website is a launching-soon placeholder, so product details aren't public yet.
The founder brings hands-on software-engineering experience from large technology and finance organisations, which matters for building reliable autonomous systems. Beyond that, the small team is still assembling, so execution is the thing to judge later.
Who else is building in this space
Funded · $125K
Cloud development environments for AI coding agents.
Pre-funding
Empowering businesses with innovative AI-driven data analytics solutions.
Funded · $125K
AI-native platform for multifamily real estate development.
Funded · $125K
Open-source SEO platform for research and analytics.
Thousands more founded every month.
Dealroom tracks new companies as they appear — founders, funding and signals, over the API or in the platform.