Basetwo Hybrid Process Digital Twin and AI Agents

Basetwo is a Canadian industrial AI company with offices in Toronto, Boston and Abu Dhabi, led by brothers Thouheed and Thamjeeth Abdul Gaffoor, both former Autodesk executives. It helps process development, tech transfer, scale-up and manufacturing teams in pharmaceuticals, personal care and specialty chemicals reduce trial-and-error experiments and anticipate deviations.

The platform ingests and cleans historical process data, then builds hybrid models that combine mass balance, kinetics and transport equations with machine learning. Engineers reuse models through a custom unit builder, simulate feed strategies and operating conditions beyond historical ranges, and use AI agents that propose experiments and return root-cause analysis. Apps and dashboards support monitoring and what-if scenarios in GMP-validated workflows.

• Data preparation: ingests and cleans historical process data and flags sensor drift.
• Hybrid modeling: combines mass balance, kinetics and transport equations with machine learning.
• Custom unit builder: turns validated models into reusable process units.
• Process simulation: tests feed strategies and operating conditions beyond historical ranges.
• AI agents: propose experiments and pilot trials and return root-cause analysis for deviations.
• Compliance-ready apps: dashboards and what-if tools in GMP-validated, Part 11 compliant workflows.

A pharmaceutical plant in Vietnam could model granulation or fermentation steps to speed tech transfer of generic drugs. A cosmetics manufacturer could optimize batch conditions for new formulations. A specialty chemicals producer could reduce off-spec batches by predicting deviations early.

Hybrid physics and ML models can predict behavior at new scales and conditions where purely data-driven models fail, reducing costly pilot runs. Agents flag data issues such as sensor drift and explain deviations. Basetwo states that it works with 6 of the top 20 pharma companies and supports more than 1,000 production workflows, with FDA 21 CFR Part 11 and SOC 2 Type II compliance.

Pharmaceutical, cosmetics and chemical producers face pressure to scale new products faster while meeting strict quality rules. Hybrid physics-informed AI is gaining ground because it needs less data than pure ML and gives explainable results, which suits regulated manufacturing growing in Southeast Asia.

Basetwo lists offices in Toronto, Boston and Abu Dhabi. It raised an 11.5 million USD Series A in January 2025, joined the AWS Generative AI Accelerator in 2025 and signed a distribution partnership with NS Solutions in Japan. Investors include Glasswing Ventures, Deloitte, Chiyoda Corporation, Shimadzu, Hub71 and Boeing.