Amously provides AI‑driven virtual metrology for semiconductor fabs, predicting wafer outcomes in real time so manufacturers can adjust process parameters without waiting for physical measurements. Its platform automates run‑to‑run control loops, continuously optimizing recipes across tools and shifts, which reduces process variation, speeds cycle time, and lessens reliance on scarce expert engineers.
Funding
Funding not disclosed
Founders
Product
Problem
Semiconductor fabs rely on physical metrology measurements that are time‑consuming, creating feedback delays that hinder real‑time process adjustments. Additionally, expert process engineers are scarce, making it difficult to maintain optimal control across advanced manufacturing lines.
Solution
Amously delivers AI‑driven virtual metrology that predicts key wafer outcomes in real time, allowing manufacturers to make control decisions without waiting for physical measurements. By estimating process outputs on‑the‑fly, the platform reduces feedback latency and supports run‑to‑run (R2R) control loops that continuously optimize production parameters. The solution is built on production‑validated models proven in high‑volume fabs, translating expert‑level decision logic into software that scales with declining workforce availability. Integration with existing fab data streams enables seamless adoption and immediate impact on yield and cycle time.
Target Audience
Primary customers are semiconductor manufacturers and fab operators seeking to improve yield, reduce cycle time, and automate process control in advanced process nodes.
Features
- Real‑time virtual metrology models that forecast wafer characteristics before physical measurement
- Automated run‑to‑run control algorithms that adjust process settings based on predicted outcomes
- Production‑validated AI models trained on historical fab data for advanced semiconductor nodes
- Compatibility with standard fab data infrastructure for easy integration into existing workflows
- Continuous learning framework that updates models as new process data become available
- Enterprise‑grade deployment with validation, governance, and ROI assurance processes