- Current focus
- Avos Lab: An operating system for trillions of AI agents.
- Based in
- San Francisco · Silicon Valley
- For investors
- Open to conversations on AI infrastructure & Avos Lab.
A builder who learned to scale systems: human, mechanical, and now intelligent.
The work began at seventeen. My first company was an education-services business that served more than 500 customers and exited in 2016. That same year I started a textile manufacturer in Bangladesh out of a single room and scaled it into a $100K+ business in five years, exporting to five countries. It was an education in operations, manufacturing, hiring, supply chain, and execution at speed.
One pattern has always driven the work: enter a hard domain, learn it deeply, build from first principles, and turn that knowledge into systems people can use. Alongside the business I built and led an applied R&D lab of seven, including professors and graduate researchers, producing five peer-reviewed publications, 480+ citations, and a pending machine-learning patent for industrial prediction. The papers are in materials science and energy storage; the patent in machine learning and data science.
I studied Applied Chemistry and Chemical Engineering at one of Bangladesh's top universities (under 2% acceptance), completed an M.S. in Applied Data Science at the University of San Diego with a 3.89 GPA, and spent three-plus years in Stanford's AI ecosystem through audited and independent study, covering deep learning, NLP, reinforcement learning, transformers, ML systems, compound AI, and self-improving agents.
Now I'm applying that foundation at Avos Lab: the context, memory, encryption, and control plane for a world where billions, eventually trillions, of AI agents operate across software, business, and human systems.
“If the last era of software was built around apps and cloud services, the next will be built around autonomous agents, and those agents will need an operating layer for context.”