The operator introduces the Cadence Third Quarter 2023 Earnings Conference Call and welcomes everyone. The call will include a question-and-answer session after prepared remarks. Richard Gu, Vice President of Investor Relations, and other executives will be speaking. The discussion will include forward-looking statements and non-GAAP measures. Anirudh Devgan, President and CEO, will be speaking next.

Cadence has reported strong results for the third quarter of 2023, exceeding their Q3 guidance and raising their outlook for the year. Design activity remains strong, driven by trends such as AI, hyperscale computing, 5G, and autonomous driving. The integration between electrical and mechanical domains, as well as the increase in 3D-IC and chiplet designs, is driving the need for AI-driven design automation. Cadence's Generative AI platform, JedAI, is experiencing growing momentum with customers and has nearly tripled in sales in the past year. Major companies such as NVIDIA, Tesla, and Broadcom have adopted Cadence's solutions and have achieved impressive quality of results.

In Q3, Cadence made progress in leveraging GenAI's LLM capabilities for chip design and renewed collaborations with major customers in various industries. The digital IC business saw 11 new wins and successful deployment of the Cadence Cerebrus GenAI solution. The Functional Verification business also had a strong quarter with 18% revenue growth, driven by increasing demand for Palladium Z2 and Protium X2 hardware platforms.

In the third quarter, Cadence's hardware business had a record performance, with almost half of the orders including both platforms. They had successful partnerships with top AI and automotive chip suppliers, as well as a major deal with a datacenter chip company. Their Custom IC business saw a 15% revenue growth and their AI-driven Virtuoso Studio solution was well-received by customers. The company also invested in their IP business and acquired Rambus PHY IP assets, leading to a record bookings quarter. Their System Design and Analysis business also showed strong growth, and they announced new AI-driven solutions for PCB design and CFD. Overall, Cadence's team showed continued innovation and execution.

In the fifth paragraph, the speaker discusses the company's strong position in the market and their commitment to helping customers with their product design. They also mention their concern for the ongoing conflict in the Middle East and their support for employees and families in the region through humanitarian aid. The speaker then hands over to John, who reports on the company's financial results for the third quarter and provides an updated outlook for fiscal 2023.

Cadence expects strong performance in the fourth quarter, with non-GAAP operating margin in the range of 41.5% to 42%, GAAP EPS in the range of $3.48 to $3.54, and non-GAAP EPS in the range of $5.07 to $5.13. They also expect to use at least 50% of their annual free cash flow to repurchase shares. The company's CFO has published a commentary document on their Investor Relations website with further analysis and reconciliations. Overall, they are on track to deliver a strong 2023, with expected revenue growth of 15%, non-GAAP operating margin of 41.75%, and non-GAAP EPS of $5.10. The company thanks their customers, partners, and employees for their support. The first question from a Needham analyst is about the company's backlog, which has increased compared to the previous quarter. The analyst asks if the backlog will continue to grow by the end of the year.

During the Q&A section of the earnings call, John Wall responds to a question about the company's strong second half bookings and raised full year revenue outlook. He mentions that their expectations for bookings in Q4 are very strong and that the timing of hardware installations in China contributed to the higher revenue in Q3. Anirudh Devgan also mentions that the company is seeing a lot of activity in AI and they plan to provide more KPIs related to AI in the future.

The company is involved in the AI design process in multiple ways, including working with system and semiconductor companies and applying AI to their own products. They have seen significant growth in revenue from their own AI products, but it is difficult to predict the revenue from the buildout of AI infrastructure. There has been a tripling of revenue from their own AI products in the past year. There has been no reclassification of revenue.

The speaker asks about the impact of regulations and the company's visibility into Chinese customers' use of EDA tools for mature versus advanced nodes. The company states that the recent changes in regulations do not have a significant impact on their business and their guidance includes the impact of these regulations. They also mention that they closely follow all U.S. regulations and that the latest changes are not material to their business. The speaker then brings up the macro conditions in the semiconductor industry and notes that recovery seems more gradual than expected in various end markets.

The auto, industrial, and enterprise service provider market is still soft, but there are signs of strong design activity and continued investment in R&D. The company is expecting a strong Q4 for their IP group and their hardware production is keeping up with demand. One of their large AI SoC customers is accelerating their chip road map to keep up with the complexity of next-generation AI compute workloads.

The speaker discusses the current state of the semiconductor industry, noting that some segments are weaker while others, such as data centers and automotive, are experiencing strong design activity. This is driven by the need for increased efficiency and productivity, as well as the use of AI. The speaker predicts that this trend will continue for the next few years, particularly in these two verticals.

Gary Mobley of Wells Fargo asks about the trend of upfront license revenue and its influence on overall growth for the next year. John Wall, CEO of the company, notes that the upfront revenue has ticked up to 15% this year due to the strength of hardware. He also mentions that the ratable and recurring part of the business is expected to have a 13% growth rate for the current year and over a three-year period. Anirudh Devgan, CTO, is asked about the commercialization of the five AI tools and their impact on baseline licensing activity. He notes that they are being carefully monitored and are currently being engaged with customers on top of existing platforms.

The adoption of AI products is still in its early stages and it will take some time for them to be fully deployed. The progress of deployment depends on the customer, with some adopting them faster than others. The top customers are fully engaged and seeing significant improvements, such as 8-10% power improvement from Cerebrus. This is equivalent to a node migration.

EDA market environment is being impacted by the current macroeconomic situation, causing some customers to be more prudent with their R&D investments. However, design activity is still strong among large customers in the big segment. Q3 had good bookings, but Q4 will provide a better idea of the market going forward.

John Wall and Jay Vleeschhouwer discuss the trend of customers purchasing add-ons during their baseline contracts and the impact of new products on contract renewals. They also discuss the role of AI in derisking schedules and the potential shift in EDA spending from the back end to the front end. Anirudh Devgan emphasizes the potential for AI to increase design activity and improve productivity and risk mitigation in the design process.

Jason is asking about the Q4 guidance and why there isn't more upside. John explains that the Q3 beat and raise was due to a conservative Q3 guide for hardware installations. The year has been raised by $10 million at the midpoint.

The speaker expects a strong quarter for their IP Silicon Solutions group, which could potentially bring upside. They also discuss backlog, which has been impacted by hardware lead times, but is expected to increase in the second half of the year. The speaker also mentions tracking annual backlog values and expects less hardware to be included in the upcoming backlog. A question is then asked about year-over-year growth.

During a recent earnings call, Vivek Arya asked Anirudh Devgan about Cadence's outlook for 2024, given that their model is 85% recurring. Devgan declined to comment, stating that they prefer to wait until the end of the year to assess their performance before making predictions. Arya also inquired about the strength of Cadence's IP business, to which John Wall stated that they are expecting a strong quarter in Q4, with a growth rate of 15-20%. Devgan added that more customers are outsourcing their IP needs, and Cadence's profitability in this area has improved in recent years.

Anirudh discussed the collaboration with Renesas and the incorporation of Generative AI and LLM into chip design. He mentioned expectations for improvements in quality, efficiency, and productivity, but noted that this is a longer-term collaboration and testing is currently underway. The timeframe for incorporating these tools into chip design is uncertain. He did not specify if this is a leading-edge design for Renesas.

Anirudh Devgan discusses the collaboration between his company and Renesas, highlighting their use of AI in the design process. He mentions that Renesas is using all of their AI tools and they are also collaborating on LLM. This collaboration is beneficial for both companies as it improves the quality of results and productivity. Additionally, it is especially useful for large companies with geographically diverse teams, as the AI tool can outperform human experts and help improve the performance of teams that may not be as strong.

AI is having a significant impact on the chip design industry. The gap between experienced and less experienced engineers is narrowing, making it easier for less experienced engineers to be productive. This not only benefits productivity and quality, but also has a workforce management benefit for large organizations like Renesas. The companies that adopt AI tools first and faster will benefit more compared to their peers. This is evident in fast-moving companies like Broadcom, NVIDIA, Tesla, and others. The workforce development benefit of AI is profound and is expected to be incremental growth for the EDA industry.

The speaker is asked about the potential for growth in the company's software renewals and add-ons over the next 12-24 months. They respond by saying that they will go at the pace of their customers and expect to see growth in AI tools over the next two contract cycles. They also mention that the pipeline for hardware is strong and that the verification group has been performing well for the past eight quarters.

In Q4, the company is expecting a strong booking quarter with a strong pipeline for hardware. They have accounted for potential shifts in customer buying patterns in their guidance. The adoption of AI tools is causing more users to make full flow decisions, leading to increased usage of the company's underlying tools. The company's AI tools, such as Cerebrus, typically run on multiple machines with multiple CPUs.

The speaker discusses the company's use of compute and underlying licenses, which can include 10 instances of Innovus for Cerebrus. They also use synthesis, place and route, sign-off, logic simulation, formal verification, and hardware in the design process. This requires more instances and runs, but it leads to better PPA. The backlog question is addressed, with the speaker mentioning a $400 million increase in current RPO in 3Q and 4Q of 2021. The changing composition of hardware and software may affect the relationship between backlog and next 12 month revenue.

The speaker discusses the growth of the company in the past year, with a large portion of the growth coming from hardware that was previously backlogged. They mention that lead times for hardware have decreased from 26-28 weeks to 8-10 weeks, and they expect this trend to continue. The speaker also mentions the upcoming Q4 and the progress made with the OpenEye acquisition and the Molecular Sciences group, expressing excitement for the future potential in this area.

The speaker discusses their company's product strategy, which consists of three layers: software products for simulation, computational hardware, and AI orchestration. They mention specific examples of each layer and how they can be applied in various industries such as EDA, chip design, and drug discovery. The recent acquisition of OpenEye adds a crucial middle layer of physics-based biological simulation to their product offerings.

The company has already implemented GPUs for biosimulation, resulting in increased speed. They have also expanded collaboration with a top pharmaceutical company for traditional and AI-based drug discovery. The future of AI adoption will involve building infrastructure, applying AI to their own products, and using AI in areas that were previously not automated, such as digital biology and life sciences. This will require a three-layer approach, with AI on top, biosimulation with OpenEye, and computational hardware. The CEO is optimistic about the future, but acknowledges that it will take time to fully develop.

The speaker discusses the potential for growth in biology and biosimulation at OpenEye, and mentions the company's disciplined financial approach. A question is asked about the operating margin and the speaker explains that recent acquisitions have had a dilutive effect. The speaker also mentions that the company continues to attract top talent and has recently undergone some restructuring.

In August, Cadence implemented a restructuring plan to align resources with their business strategy, resulting in $12 million in costs for severance and termination benefits. This was seen as a necessary step for the company's growth in the following year. The company's CEO expressed pride in their culture and focus on sustainable innovation, and thanked customers, partners, and investors. The conference call has now concluded.

This summary was generated with AI and may contain some inaccuracies.