Exigent provides a physics‑infused neural network platform that accelerates lithography pattern simulation, delivering full‑chip results in seconds on GPU/TPU clusters while preserving sub‑nanometer accuracy. The platform also includes a factory‑general‑intelligence engine that ingests sensor data to recommend process adjustments, predict equipment wear, and integrates via RESTful and gRPC APIs with CAD, MES, and ERP systems, offered as cloud SaaS or on‑premise Kubernetes.
Funding
Funding not disclosed
Founders
Product
Problem
Semiconductor and advanced manufacturing facilities rely on high‑resolution lithography simulations that are computationally intensive and often require days of compute time, limiting throughput and design iteration speed. Additionally, many factories lack integrated AI tools to continuously optimize process parameters across complex production lines, leading to sub‑optimal yields and higher operational costs.
Solution
EXIGENT delivers a physics‑infused neural network platform that accelerates lithography pattern simulation by orders of magnitude while preserving physical fidelity. The core simulator runs on GPU/TPU clusters and produces near‑real‑time results, enabling designers to evaluate mask layouts within minutes instead of days. Complementary factory‑general‑intelligence modules ingest sensor streams, production schedules, and quality metrics to generate prescriptive adjustments for equipment settings, material usage, and workflow sequencing. All components are exposed through RESTful and gRPC APIs, allowing seamless integration with existing CAD, MES, and ERP systems. The platform is offered as a cloud‑hosted SaaS solution with optional on‑premise deployment for highly regulated environments, providing continuous model updates and automated performance monitoring.
Features
- Physics‑infused neural network architecture that embeds Maxwell’s equations and diffraction models directly into the learning layer
- GPU/TPU‑optimized inference engine delivering lithography simulations in under 5 seconds for full‑chip layouts
- Automatic mesh refinement and adaptive sampling to maintain sub‑nanometer accuracy across varying pattern densities
- Real‑time API endpoints for CAD tool integration, supporting industry formats such as GDSII, OASIS, and OPC
- Factory‑general‑intelligence engine that correlates process sensor data with simulation outcomes to recommend optimal process windows
- Predictive maintenance module that forecasts equipment wear based on simulation‑derived stress patterns and historical logs
- Role‑based access control and end‑to‑end encryption compliant with ISO 27001 and IEC 62443 standards
- Scalable SaaS deployment with optional on‑premise Kubernetes bundle for data‑sensitive facilities