^※ Gauss Labs is an AI software startup based in Silicon Valley, USA, founded in August 2020 with capital investment from SK hynix. In November, the company will release PANOPTES VM (Virtual Metrology), an AI-based software solution developed by the company that predicts process performance on unmeasured wafers using sensor data from semiconductor equipment and data measured through wafers. Through this, Gauss Labs is expected to be in a unique position to spearhead industrial AI innovation to the next level.

Since 2021, Gauss Labs has been offering internships to foster global talents for developing cutting-edge industrial AI solutions, and it continues to provide opportunities for growth by building on AI knowledge gained so far and applying it to projects carried out on actual industrial sites.

Positions across five roles are available for this internship opportunity: Applied Scientist (Machine Learning & Data Science), Applied Scientist (Computer Vision & Image Processing), AI Engineer, Computer Vision & Image Processing Engineer, and Software Engineer. In addition to an interview, candidates may be asked to take a coding test.

Applicants can apply until October 17 (Mon) through the Gauss Labs website (https://jobs.lever.co/gausslabs). Successful applicants will be notified individually in November. Finalists will work full-time in Seoul for 9 weeks from the end of December this year, with the option of continuing to participate in the project on a part-time basis beginning in March and during the semester. After the project is completed, those who have displayed excellent performance will be given the opportunity to intern in Silicon Valley for around 8 weeks. All lodging expenses and return flights from Seoul will be fully covered by Gauss Labs.

Mike Young-Han Kim, CEO of Gauss Labs, said, “The ‘Grow with Gauss’ global internship allows participants to gain hands-on experience working on real-world projects in the field of industrial AI. It offers infinite growth potential through mentoring from Gauss Labs members who are equipped with the world’s best AI technological skills.”

The post AI Startup Gauss Labs Recruits for Global Internship Program first appeared on SK hynix Newsroom.

Image Download

Artificial intelligence (AI) has long been a fascination of popular culture. Often depicted in television and film, AI companions range from the empathetic and friendly Baymax of Big Hero Six to the conniving and humanoid Eva of Ex Machina. But, at least for now, how AI is integrated into our daily lives looks a little bit different.

Many people are aware of its use within the fields of security, healthcare, and marketing but AI technologies may soon become even more commonplace. Industries all over the world are making the move to embrace digital transformation—and AI stands to be a large part of it. Businesses must not only incorporate existing technologies into their service models, but also look forward. According to a 2021 Gartner survey on AI investment, more than 50% of technology providers report significant customer adoption of AI-enabled products and services. This demonstrates a clear need for businesses to embrace technological innovations like AI to keep on top of the growing demand for simple and efficient digital service solutions.

Advanced technologies like AI rely on hardware, especially for logic and memory functions. As such, the trend in AI-enabled digital transformation (DX) could represent the biggest opportunity for semiconductor companies in the last decade. In fact, growth in the semiconductor market in the area of AI is expected to be five times that of the growth in all other areas by 2025.

Matchmaking Automation and Intelligence

Traditionally, automatons were largely stationary manufacturing machines and often could be found performing highly repetitive or dangerous tasks in factory lines to increase production speed, reduce accident rates, and eliminate human error. Even today, a large portion of the work completed by autonomous machinery still takes place on a production line. Due to the routine and repetitive nature of many manufacturing tasks, this kind of automated machinery does not need to evaluate external variables or revise its own protocols and is limited to operating a series of simple tasks.

But, with advancements in AI and machine-learning, there is increased potential for automated machines to become more mobile, resourceful, and intelligent. These improvements are an important part of making digital automation relevant to industries outside of manufacturing.

Figure 1. Global investment is AI across various disciplines in 2019 and 2020.

Image Download

Enabling automated technologies to support the more complex, non-routine service needs of a variety of industries means countless more businesses can partake in modern day digital transformation in a more impactful way.

The Nuts and Bolts

Artificial intelligence describes a machine or software application’s ability to reason, learn, and act in a manner similar to human thought, so much so that parallels are often drawn between the human mind and AI systems. Just as synapses are essential to the transmission of nervous impulses within our brains, semiconductor memory stores and processes information as a part of AI systems for our mechanical counterparts.

Beyond its current applications in Big Data analytics and facial recognition software, AI is critical to enabling the next generation of automated solutions to problem solve, reason, and evaluate. This important evolution will expand the application of automated technologies, allowing for increased participation in intricate, multi-faceted problem-solving tasks.

In the first quarter of 2021, USD $1.7 billion was invested into artificial intelligence and machine learning chip makers, demonstrating clear means and a strong incentive for innovation in the AI semiconductor space. Semiconductors for use in AI systems must be designed specifically to handle complex workloads and parallel operations. These AI systems have to process incredible amounts of data quickly and accurately to understand and respond to environmental and stimulus changes.

Fortunately, industry leaders such as SK hynix are working diligently on memory and storage solutions that can support the demanding requirement of AI technologies. Advancements such as SK hynix’s 176-layer 4D NAND flash and DDR5 DRAM stand to play a large role in defining and supporting future AI and machine learning offerings.

Figure 2. SK hynix’s 176-layer 4D NAND Flash

Image Download

Figure 3. SK hynix’s DDR5 DRAM

Image Download

As AI continues to permeate business practices around the world, increased opportunities will arise in a number of areas including AI-optimized storage systems, infrastructure for data centers, programmable switches and high-speed interconnect. In tandem, the market is likely to begin seeking more end-to-end solutions from semiconductor manufacturers to satisfy business-wide digital transformation.

Supercharging Digital Transformation

Figure 4. The six digital platforms leading SK hynix’s digital transformation

Image Download

Digital transformation has been an important part of the corporate zeitgeist in recent years. This adoption of digital technology to improve business processes and deliver increased value is an important part of maintaining a business’ competitive edge in 2021 and beyond. Incorporating AI technologies offers an avenue for businesses to do more than simply improve work efficiency.

RPA, short for robotic process automation, sees businesses using automated machinery to complete routine tasks. Reminiscent of initial automatons, this is one way to undergo digital transformation. Powered with AI, the applications of RPA become much broader. If an unexpected event were to occur, a simple software robot would not be able to adjust and continue its work. To manage irregularities, it must be able to evaluate and make judgements, course correcting as it goes.

Many companies shifted to including more RPA in their service models in 2020 as a response to the COVID-19 pandemic following an increase in consumer demand for socially distanced services. The business demand for automated integrations skyrocketed as a result with year-over-year orders for automated machines in life sciences, food and consumer goods, and automotive growing 69%, 56%, and 39% respectively.

Figure 5. Percentage of Executive Survey respondents increasing the pace
of their AI/Big Data investment.

Image Download

AI creates solutions that can accomplish more complex goals, meaning the level of human intervention required is decreased and the impact of digital transformation amplified. This means the complex problems often encountered in service-focused, customer-facing tasks can be resolved independently by AI solutions, answering the important social and cultural trends that surround the pandemic.

As a part of its AI-driven efforts, SK hynix, a major player in the information and communication technology (ICT) industry, recently established Gauss Labs Inc.—an industrial AI solutions provider. Gauss Labs Inc. is the first AI company from SK Group to lead manufacturing innovation, marking a clear investment in internalizing AI practices by the forward-thinking ICT giant.

Figure 6. (From the left) Brendon Choe (Head of Computer Vision), Sunghee Yun (Head of R&D), Mike Young-Han Kim (CEO), Taeyoung Park (Chief of Staff), and Moosung Kim (Head of PMO) at Gauss Labs Inc.

Image Download

According to Gartner, one-third of technology providers with AI plans said they would invest USD $1 million or more into the technology over the next two years. Adopting automated solutions as a part of digital transformation will inevitably lead to an explosion in data volume and an increased demand for Big Data analysis and machine learning. As a result, corporate spending in Big Data and analytics is also on the rise with European spending in the area forecast to reach USD $50 billion this year. A parallel increase in the demand for data centers and the edge is likely—with some experts predicting a combined market value topping USD $18 billion by 2025. This growth is bound to spark additional innovation in the next generation of memory solutions.

Integrating AI technologies into digital transformation is an important part of the fourth wave of digital revolution in-step with the rise of robotics, machine learning, and demands for socially distanced solutions.

What automation began; AI supported by semiconductor memory will innovate—creating digital solutions primed to thrive in the face of the future.

The post Cracking the Code: Leveraging AI Technology in Digital Transformation first appeared on SK hynix Newsroom.

Now, ten months later, the company is preparing for another leap. With projects now on track and the company properly set up, Gauss Labs plans to double the number of its employees by the end of this year. Newsroom sat down with the Gaussians to learn more about the company’s projects, what kind of talents they are looking for, and the blueprint for the company’s future.

Taking the Lead in the Blue Ocean of Industrial AI

Industrial AI, unlike service AI, has no dominant player in the market and is considered a ‘blue ocean’ of diverse opportunities. Among various opportunities in the industrial AI field, Gauss Labs chose to take on “semiconductor,” given the enormous economic impact when AI is applied to the semiconductor sector and SK hynix’s need to fortify its AI capabilities.

(From the left) Brendon Choe (Head of Computer Vision), Sunghee Yun (Head of R&D), Mike Young-Han Kim (CEO), Taeyoung Park (Chief of Staff), and Moosung Kim (Head of PMO)

Image Download

CEO Mike Young-Han Kim stated that the semiconductor industry is not only one of the most important sectors globally but also has a significant impact on the Korean economy.

“A combination of AI and semiconductor technology will bring significant improvement in efficiency amid growing complexity stemming from continued scaling efforts,” he said.

Market observers forecast the semiconductor sector to generate values worth KRW 100 trillion in the industrial AI market in the mid-to-long term. AI technology acquired from the semiconductor manufacturing process is expected to be easily applied to other industries for quick expansion– another reason why Gauss Labs chose the semiconductor sector as its first project.

Mike stated that “the semiconductor industry is called the ‘center of precision manufacturing,’ as it combines various fields including chemical processing, mechanical processing, and even optics processing.
“Success in addressing various challenges in this area will allow us to expand our scope to other industries. Our goal is to engage with various global manufacturing companies with our specialty in industrial AI. We start with SK hynix and will expand to other manufacturing affiliates of SK Group next.”

Gauss Labs has been developing AI solutions aimed at maximizing production efficiency using the massive amount of data generated at SK hynix’s production sites. SK hynix wishes to make the overall semiconductor production process more intelligent and optimized across all procedures including process management, yield prediction, equipment repair and maintenance, materials measurement, and defect testing and prevention.

Raising Yields and Reducing Costs With AI-combined Process Monitoring Solutions

To address the challenges facing semiconductor manufacturing, Gauss Labs is focusing on developing a process monitoring solution using AI. Monitoring technology is essential to all manufacturing processes, not just semiconductors. This is because manufacturing processes are required to reliably yield output of consistent quality and be robust enough to minimize the impact of an unpredicted change. Semiconductors, in particular, require extremely precise manufacturing technology that does not allow even a slight deviation.

“The manufacturing process comprises hundreds of complicated and intricate processes, which easily lead to potential defects. Therefore, engineers monitor the process or equipment they are responsible for in-person or in front of their PCs. Gauss Labs is currently developing three solutions to enable engineers to conduct monitoring accurately and conveniently using AI technology. Our goal is to develop an AI assistant that helps engineers make smart decisions, not to replace them,” – Moo-sung Kim, Head of PMO at Gauss Labs.

Gauss Labs is currently developing three solutions; Automatic Image Metrology for Semiconductor (AIMS), Virtual Metrology (VM), and Statistical Process Control (SPC).

Image Download

AIMS is a solution that innovates metrology based on image data. In semiconductor manufacturing, the size of the circle and the level of damage is checked using electron microscopes to find defects. SK hynix generates millions of image data every day. AIMS uses computer vision (technology that builds human visual systems using certain algorithms) to enhance image quality and automatically take measurements, enabling faster and more accurate measurement. This will help semiconductor companies save costs since measuring equipment is extremely expensive.

Virtual Metrology (VM) is a solution that enables measuring without actually measuring. It uses equipment’s sensor data, which is generated when wafers are processed, and provides engineers an estimated measurement. This is especially useful for monitoring the condition of equipment and process. Applying a VM solution can save costs on equipment investments and maximize the efficiency of existing measuring equipment. Gauss Labs’ goal is to enhance the capability of VM so that it can completely replace actual measuring.

Statistical Process Control (SPC) is a monitoring solution that analyzes the probability of a problem occurring in each process or equipment with the help of machine learning and quickly lets engineers identify the cause of the problem. This root cause analysis can minimize simple repetitive tasks, significantly reducing equipment downtime (time during which a machine is not running) and allowing engineers to focus on solving core issues instead.

Gauss Labs expects the three solutions to generate economic effects worth hundreds of billions of KRW when fully applied in the overall manufacturing process.

“Our ultimate goal is to build an industrial AI platform by integrating all the solutions we develop. Applying the same solutions to different industries won’t be easy, but we will be able to easily apply the underlying technologies that make up the applications. We plan to build an open platform with the underlying technologies so that SK hynix and other semiconductor equipment, cloud service, and software companies can also freely participate and further develop, thus achieving innovation in industrial AI,” – Sunghee Yun, the Head of R&D.

How the Gauss Labs Works

Gauss Labs consists of two parts: the R&D organization that studies and develops AI solutions and the Program Management Office (PMO) that commercializes the AI solutions. The company has recruited the best experts in each area to strengthen its competency.

Sunghee Yun joined as Head of R&D. He comes from Amazon, where he built his expertise in AI technology through various fields including e-commerce after studying machine learning and convex optimization at school.

Sunghee Yun (Head of R&D)

Image Download

The PMO organization is led by Moosung Kim, an experienced engineer who built his career at SK hynix and Intel. Kim is an expert in not only semiconductor manufacturing and equipment, but also AI through his experiences as a citizen data scientist.

Professors from prestigious Korean universities are also actively participating in Gauss Labs projects part-time in the fields of machine learning, statistics, control, vision, and scheduling. Gauss Labs is closely collaborating with various academics by attending related seminars by scholars at home and abroad; the company also plans to expand knowledge exchange programs in the coming months.

The R&D organization is divided into four teams in accordance with the technology they cover: CV/IP Team responsible for computer vision and image processing, ML/DS Team in charge of machine learning and data science, Operations Research & Engineering Optimization (OREO) Team, which focuses on planning and scheduling of the manufacturing process, and Software Architecture & Platform (SWAP) Team for development of software platform. Together, they analyze issues based on an understanding of the domain knowledge required to solve industrial AI issues and conduct research, development, and realization of AI algorithms using AI knowledge, intuition, and experience. The R&D organization also analyzes the latest AI technology trends to build a medium-to-long-term technology roadmap for Gauss Labs.

Moosung Kim, Head of PMO

Image Download

The PMO organization designs products that customers need and manages projects. It must clearly define problems and issues at the manufacturing sites and communicate with AI engineers by converting the problems into mathematical language. As project manager and product owner, PMO must also set the direction for customers’ strategy in the long term.

Given the nature of industrial AI problems, even a single project requires a solution that integrates several technologies. Therefore, Gaussians are not bound to their teams but may instead work cross-functionally with other teams as a member of the same project under the matrix system for maximum synergy.

Gauss Labs also works closely with various relevant organizations, including SK hynix’s Manufacturing/Technology, Revolutionary Technology Center, and DT. Particularly, it’s working closely with SK hynix’s AI/DT Solution under Manufacturing/Technology and Data Architecture Team to jointly design the blueprint for a smart fab. To perform this project, they share data and domain knowledge needed in the development, monitor whether the solutions under development are appropriately applied at actual sites, and discuss key projects through regular meetings twice a week. Quarterly training that includes a fab tour is offered to Gauss Labs AI engineers to help them gain domain knowledge of the semiconductor industry.

“AI experts are not necessarily familiar with the domain, and domain experts don’t necessarily have a good understanding of how AI works. But in reality, you should know both to effectively apply the domain knowledge in building AI algorithms. We knew this would be a challenge for a team composed of only AI engineers. Furthermore, because Gauss Labs is not an internal team of SK hynix, it was very difficult to work closely together. So SK hynix created a counterpart team when Gauss Labs was born. Although SK hynix is our customer, we consider them as one team as we work together.” – Moosung Kim

Looking for new Gaussians to Join Our Journey

Gauss Labs is currently composed of 22 members and plans to more than double in size to 50 between the US headquarters and its Seoul office by the end of this year. Recruitment in the US, which has been inactive due to COVID restrictions, will begin in earnest starting June.

What capabilities and qualifications would Gauss Labs want from job candidates? And what would be a good reason for AI engineers to choose Gauss Labs, among others? CEO Mike shared his ideas and vision with Newsroom.

Mike Young-Han Kim, Gauss Labs CEO

Image Download

Q. What makes the combination of AI and semiconductors attractive?
There is no other industry that generates as much data as the semiconductor industry. Being able to deal with real data generated from actual manufacturing sites will be a valuable experience for any AI engineer. Also, the opportunity to collaborate with frontline employees on real problems and data on-site is not easy to come by if you’re working for any startup. You will find several issues that are hard to address but worth exploring from the semiconductor manufacturing process. Another factor is that you will be able to achieve enormous economic impact through the solution you help create.

Q. What are the capabilities and qualities required from a Gauss Labs engineer?
Gauss Labs is walking a road not taken before and must deal with problems that are unsolvable with existing AI technology. We’re the ones who will define the problems to solve, creatively apply new forms of algorithms and insights to the problem’s core, and consider technologies from multiple angles. In short, we need engineers who are willing to face various challenges and bring innovative solutions.

Image Download

Gauss Labs refers to its principles when hiring new talent. The ‘Gaussian Principles’ define seven core values; Focus on What and Why, Highest Standards, Learning with Growth Mindset, Readiness for Deep Dive, Ownership and Accountability, Team Spirit, and Open Discussion and Communication. We are looking for talents who manifest these principles.

Q. What benefits would a Gauss Labs AI engineer gain?
The two hottest keywords today are AI and semiconductors. These two are the most important technologies that could change the fortune of a nation and provide the future growth engine in the era of the fourth industrial revolution. So anyone who has built a career at Gauss Labs, dealing with AI and semiconductors, will eventually mature into one of the most sought-after talents in the world. In addition, such experience will not limit the engineer to the semiconductor industry but will provide a foundation that will allow the engineer to expand to other industries.

Q. Any message for AI talent who wish to join Gauss Labs in the near future?
There’s no Google, Microsoft, or Amazon in the industrial AI area. It will be a tough journey, but the opportunity to become the world’s top player exists. Don’t miss the chance to advance in this field. Gauss Labs already has many talented engineers working to becoming the best in industrial AI, and we’re waiting for more to join our journey. There is much to learn here as an AI expert. We look forward to working with new AI talent with us soon.

The post Aiming for the Top in Industrial AI, SK’s First AI Company Gauss Labs first appeared on SK hynix Newsroom.

Seoul, September 22, 2020

SK hynix Inc. (www.skhynix.com) announced that Gauss Labs Inc. (www.gauss-labs.com) was founded to lead manufacturing innovation through industrial AI solutions.

Gauss Labs will start its Korean office later this month right after having established its headquarters in Silicon Valley, U.S. last month. SK hynix is the sole investor of its initial capital of $55 million through 2022.

SK Group has been promoting various types of AI businesses with its affiliates. “We should utilize innovative technologies such as AI and digital transformation (DT) to create social value, expand our market, and help our customers pursue happiness,” said Mr. Tae-won Chey, the Chairman of SK Group at SK Icheon Forum in August last year. “If we fail to utilize innovative technologies, we cannot guarantee the future of SK.” Gauss Labs is the first independent corporation from SK Group specializing in AI.

Gauss Labs initially aims to revolutionize semiconductor manufacturing through innovative AI technologies by utilizing an enormous amount of data generated at SK hynix’s manufacturing sites. Through these technologies, SK hynix plans to make its overall manufacturing operations, including process control, yield prediction, equipment maintenance, defect inspection and fault prevention, more efficient and more intelligent.

Young-Han Kim, a world-renowned data scientist, who is a tenured professor of the University of California, San Diego (UCSD) and a Fellow of the Institute of Electrical and Electronics Engineers (IEEE), will serve as CEO of Gauss Labs. Kim recently worked as a Data Research Fellow for SK hynix and is considered the best person to lead Gauss Labs’ growth with his technical expertise and global network.

Gauss Labs has also named Dr. Sunghee Yun from Amazon as its Head of R&D. Dr. Yun is an AI and optimization expert who has led several high-profile projects in various industries such as semiconductor and e-commerce. He is a former student of Professor Stephen Boyd at Stanford University, one of the most distinguished authorities in machine learning (ML) and convex optimization.

Gauss Labs plans to expand its business to global customers as well as SK Group affiliates related to energy, bio and other manufacturing industries. AI services have been mainly based on B2C before, but recently AI is being gradually introduced to the manufacturing industry for production troubleshooting and cost reduction.

“Gauss Labs is actively seeking the best players in AI. We plan to recruit both in our U.S. headquarters and Korean office, with the goal of growing the company to 200 scientists and engineers by 2025,” said Kim. “By working on the right problem at the right moment, I am confident that Gauss Labs will become a world-class industrial AI powerhouse with exceptional talents like Carl Friedrich Gauss.”

Gauss Labs is currently recruiting global AI talents from the industry and academia through its website.

*Gauss Labs homepage : www.gauss-labs.com
*Gauss Labs recruiting e-mail address : gausslabs-recruit@sk.com

About SK hynix Inc.

SK hynix Inc., headquartered in Korea, is the world’s top tier semiconductor supplier offering Dynamic Random Access Memory chips (“DRAM”), flash memory chips (“NAND flash”) and CMOS Image Sensors (“CIS”) for a wide range of distinguished customers globally. The Company’s shares are traded on the Korea Exchange, and the Global Depository shares are listed on the Luxemburg Stock Exchange. Further information about SK hynix is available at www.skhynix.com, news.skhynix.com.

Media Contact

SK hynix Inc.
Global Public Relations

Technical Leader
Sejin Julia Yoo
E-Mail: global_newsroom@skhynix.com

Technical Leader
Jaehwan Kevin Kim
E-Mail: global_newsroom@skhynix.com

Technical Leader
Eun Suk Yixi Lee
E-Mail: global_newsroom@skhynix.com

The post SK hynix Invests in AI Specialized Company ‘Gauss Labs Inc.’ to Lead SK’s Future Value Creation first appeared on SK hynix Newsroom.