Hardware Podcasts

How AI is Changing Every Aspect of EDA

EE Times Current · 26 Sept 2024 · 31m

AI in practice: 4 real uses of AI described

What this episode covers

Satish Kumar Balasubrahmanian, who leads product management for custom IC verification at Siemens EDA, explains how the company applies 'Verifiable AI' to accelerate semiconductor design and verification. He details the evolution from brute-force Monte Carlo simulations to AI-driven tools like Solido, which reduce simulation counts by 100x while maintaining high-sigma accuracy for mission-critical applications like automotive. Balasubrahmanian also discusses the integration of generative AI and LLMs to leverage historical design data, improve workforce training, and automate knowledge retrieval, arguing that these tools must be verifiable and robust to be production-ready.

AI in practice

Where this conversation describes AI being used to build or deliver a product. Part of the AI in practice tracker.

Engineering & simulationSolido High Sigma Monte Carloless than 10,000 simulations (vs 1,000,000+)

AI modeling reduced the number of simulations required to achieve 5-6 sigma confidence from over 100,000 to less than 10,000, providing the same accuracy as brute force methods with significantly fewer resources.

“So what, Soledo came up with this pseudo-high Sigma Monte Carlo. And this was the solution that used AI modeling in which we were able to give the same confidence of 5 to 6 sigma of root force, but with far less assimilations.”

Speaker 2 · Hear it at 15:49

Engineering & simulationCalibre

Machine learning techniques are used to detect hotspots in lithography much faster than traditional methods, optimizing resource prediction and speeding up the sign-off process.

“In caliber, caliber, semi-caliber, little tools. We use machine learning techniques to actually figure out, for example, where is the hotspot? How can you detect the hotspot in a much more fast away rather than running rather than spending too much too much time in doing it in a traditional way?”

Speaker 2 · Hear it at 15:49

Engineering & simulationQuestas platform33 X reduction on time

AI techniques are applied to functional verification to increase coverage and reduce the time spent achieving near 100% verification coverage by up to 33 times.

“The 2nd one is going to be on the functional verification, which is a questa platform. What we do is we actually increase the verification coverage … With AI, we are able to use different techniques to actually reduce the time that's spent to get almost 100% coverage, almost up to 33 X reduction on time.”

Speaker 2 · Hear it at 10:20

Engineering & simulationVelodrome platform50% reduction in compiled time

AI is used in emulation and prototyping to reduce compilation time by 50% by learning from previous compilations and adjusting hardware simulation time on the fly.

“So we have in emulation of prototyping, we have a velodge platform that actually uses AI to actually reduce the compiled time of front 50% reduction in the compiled time. That's a huge, huge, I would say, savings that we do. And we have a lot of ways of doing that. We learn from previous previous compiled.”

Speaker 2 · Hear it at 10:20

Most relevant to

large enterprises

Topics

Mentioned

Companies
Siemens EDA, Solido Design Automation, Synopsys
Products
Solido High Sigma Monte Carlo, Caliber, Questasim, Veloce, ChatGPT
Also
Python

Related episodes from other shows

Other shows on the same ground, closest first.

  • OnTrack · 58m

    Inside Simberian's AI-Driven Design Automation

    Yuriy Shlepnev, founder of Simberian, discusses the integration of Large Language Models (LLMs) into signal integrity (SI) workflows via Simbeor Commander. He contrasts this agentic approach with previous machine learning attempts that relied on training data, arguing that LLMs are better suited for orchestrating deterministic solvers than for optimization or prediction. Shlepnev demonstrates how the AI agent can autonomously create stackups, import complex data from PDFs, and compare designs, effectively acting as an engineer that calls the underlying simulation tools.

    AI & machine learningPCB design & layout
    6 chapters

  • Moore's Lobby · 1h 01m

    Synopsys' Stelios Diamantidis Is Bringing the Magic of AI To Chip Design

    Stelios Diamantidis, Distinguished Architect at Synopsys, discusses the evolution of chip design from manual processes to AI-driven optimization. He recounts his career journey, including founding two EDA startups focused on verification and Silicon Lifecycle Management (SLM), before joining Synopsys to lead the development of DSO.ai. Diamantidis explains how reinforcement learning is used to solve intractable physical design problems, optimizing power, performance, and area (PPA) for diverse verticals like CPUs, GPUs, and automotive chips.

    AI & machine learningEDA & CAD tools
    8 chapters

  • The Amp Hour · 1h 04m

    Leading Edge Analog with Joren Vaes

    Joren Vaes, a high-speed design engineer at SOFICS, explains the complexities of designing analog IP blocks for leading-edge semiconductor nodes like TSMC's 2 nm process. He details how designers navigate massive Process Design Kits (PDKs) with thousands of rules, utilizing advanced simulation tools from Cadence, Synopsys, and Siemens to manage parasitics across 18+ metal layers. The conversation covers the transition from FinFET to nanosheet architectures and the use of on-chip passive components, such as transformers, to achieve impedance matching at frequencies up to 50 GHz. Vaes also describes the rigorous verification workflow, including Design Rule Checks (DRC), Layout Versus Schematic (LVS), and parasitic extraction, which are critical before tape-out to millions of dollars in mask costs.

    PCB design & layoutChips, FPGAs & semiconductors
    9 chapters

  • The Amp Hour · 1h 07m

    Intelligent Routing with Sergiy Nesterenko

    Sergiy Nesterenko, CEO of Quilter.ai, explains how his company uses generative AI and parallel cloud compute to automate PCB layout from schematics. He discusses the company's pivot from hardware sensors to software, drawing on his experience at SpaceX where manual routing became a critical path bottleneck. Nesterenko details how Quilter handles constraints, validates designs using built-in Maxwell equation solvers, and targets specific fabs like JLCPCB and Oshpark to ensure manufacturability.

    EDA & CAD toolsAI & machine learning
    7 chapters

  • OnTrack · 47m

    EMC Testing in the Age of AI: Why Respins Aren't Going Away

    Andy Eadie, founder of Engentica and owner of EMC FastPass, explains why electromagnetic compatibility (EMC) testing pass rates often fall below 50% and argues against budgeting for inevitable respins. He demonstrates Engentica, an AI-powered design co-pilot that ingests schematics and layouts to run compliance-focused checks and answer natural language queries about design risks. Eadie discusses the limitations of fully generative AI in hardware design, noting that simulation remains essential for signal and power integrity, while his tool focuses on scaling domain expertise and catching predictable errors early.

    AI & machine learningPCB design & layout
    7 chapters

The summary, chapters and topics are generated from the episode’s audio by this directory, not written by EE Times Current. The episode itself belongs to its publisher.