NextFlex AI Webinar Series
AI for Hybrid Electronics – Workshop Outcomes & What’s Next
At the webinar on October 29, we unpacked the key takeaways from the AI Workshop and shared how NextFlex is advancing AI adoption in hybrid electronics through upcoming initiatives and industry engagement.
Watch the Recording:
View the Slides:
AI for Hybrid Electronics – Workshop Outcomes & What’s Next
Accelerating Hybrid Electronics R&D with AI Automation – Presented by Object Tech
Hybrid Electronics development demands precision across materials, printing, and process control — yet traditional methods rely heavily on manual inspection and trial-and-error optimization. On November 12, Dr. Jiaojiao Li, Co-Founder & CPO of Object, introduced how Agentic AI tools can automate critical steps in R&D, from defect inspection and classification to critical dimension (CD) metrology, anomaly detection, experiment optimization, and property prediction.
Drawing on Object’s recent pilot with flexible hybrid electronics program, Jiaojiao shared how domain-specific AI tools, models and agents can help scientists and engineers predict outcomes, identify defects faster, and optimize processes in real time. Attendees gained insight into the next generation of AI-powered R&D workflows and how these tools can potentially transform productivity and reliability across the hybrid electronics ecosystem.
Watch the Recording:
AI Opportunities in Electronics Manufacturing and Reliability
Miniaturization and heterogeneous integration technologies are increasing the complexity of semiconductor packaging, making AI essential for optimizing design, manufacturing, and reliability. On December 3, Professor Pradeep Lall gave insights into core concepts and practical applications, supported by packaging case studies that demonstrate how AI can address critical challenges.
The presentation included selected AI models and their application in analyzing process data for anomaly detection, root cause analysis, and real-time adjustment of process parameters. This webinar recording is particularly suited for packaging engineers, process specialists, reliability analysts, and technical managers who are looking to leverage AI for faster time-to-market, improved yield, and enhanced product reliability.
Watch the Recording:
Harnessing the Power of AI in FHE Design – Understanding Challenges and Unlocking Opportunities
The role of artificial intelligence (AI) in electronics design has become increasingly important due to a combination of factors, including a workforce crunch, the explosion of complexity and cost, and tighter delivery schedules. This session explored the significance of AI in design considering these factors and highlights the benefits it brings to address the challenges faced by the industry. The webinar also addressed risks to deploying AI technology in a design process and covered practical ways that AI can be applied to design for assistive development, design optimization through deep learning, and generative processes that address highly complex problems.
Watch the Recording:
Closing the Loop: AI-Driven Predictive Modeling for Printed Electronics
Printed electronics manufacturing faces a unique set of challenges, material variability, substrate compliance, and complex multi-physics interactions, that strain the limits of conventional process control. Achieving consistent, high-quality output requires moving beyond reactive inspection toward predictive, closed-loop systems that can sense, model, and correct in near real time.
Dr. Ben Davaji’s group at Northeastern University has been actively exploring how AI and machine learning can help realize that vision across real printing hardware and workflows. This talk presents an ongoing effort to develop an end-to-end AI-driven framework, studied across multiple printing platforms and print heads. Multi-modal sensing, combining ultrasound, vision, and acoustic emission, provides the real-time observability needed to feed rich datasets into digital twin models and ML/AI training pipelines. These data streams drive models trained and evaluated for print proximity correction, electrical impedance adjustment, and process parameter tuning. A computer-vision-based, process-aware slicing approach further extends the closed-loop concept upstream into design, bridging ECAD and MCAD environments to generate print-ready files that anticipate known process behaviors.
Throughout the process, Dr. Davaji’s group employs quantitative benchmarks, spanning dimensional accuracy, impedance prediction, and process yield, to assess where AI meaningfully advances outcomes and where open questions remain. Results are drawn from experimental studies across multiple process conditions and printing platforms, reflecting both the promise and the practical challenges of dataset creation, model generalization, and workflow integration. Looking ahead, the implications of AI for printed electronics extend well beyond process control, from accelerating materials discovery and enabling adaptive design to supporting supply chain resilience and lowering barriers to low-volume manufacturing.
Watch the Recording: