About the Speaker

Chris leads AI and Machine Learning at Thalia Design Automation where he has worked for the last fourteen years helping to establish Thalia as a world leader in analog migration. He has developed many advanced algorithms to solve the complex challenges in analog design migration. This work covers supervised and unsupervised learning, optimisation, statistical modelling and graph-theory applied to build scalable, high-impact solutions used in production environments. He began his career in academia with a PhD in Operations Research. He developed new methodologies for large scale mathematical models and was awarded the Goodeve medal for his research.

Structured AMS migration

Overview

Analog migration is becoming more common as technologies evolve. However, manual migration is slow with few senior engineers available. Adopting a structured approach combining AI and engineering know-how can reduce migration times by up to 50%. Mapping devices between source and target technologies ensures similar electrical behaviour; rules-based circuit migration and optimisation by identifying the key devices to meet PPA requirements; layout migration that retains as much of the proven source layout intelligence applied as a structured migration can reduce migrations cycles and reduce reliance on limit engineering resources. This is demonstrated with the use of Thalia’s Amalia migration platform.

Key Points

• Manual analog migration is slow and too few senior engineers available
• A structured approach combining AI and engineering expertise can reduce times by up to 50%
• Such and approach is demonstrated by Thalia’s Amalia analog migration platform