How To Evaluate AI Software

Buying off-the-shelf AI (Artificial Intelligence) software is a good first step for those companies that are new to the technology. There should be little need to make investments in technical infrastructure or to hire expensive data sciences. There will also be the benefit of getting a solution that has been tested by other customers. For the most part, there should be confidence in the accuracy levels as the algorithms will probably be implemented properly.

But there is a nagging issue: there are many AI applications on the market and it is extremely difficult to determine which is the best option. After all, it seems that most tech vendors are extolling their AI capabilities as a way to stand out from the crowd.

Then what are some factors to consider when evaluating a new solution? Let’s take a look at the following:

Data Connectors: AI is useless without data. It is the fuel for the insights. 

But when it comes to a new AI solution, it can be tough to find the right sources of data, wrangle it and integrate it. Thus, when evaluating an application, you need to make sure that there are ways to handle this process. 

“The most complex task in an AI solution is not to implement the machine learning algorithm anymore—this is usually available as a set of functions in every tool—but to collect the data,” said Rosaria Silipo, a Ph.D. and a principal data scientist at KNIME. “That is, to connect to a variety of data sources, on premise, on the web, or on the cloud, and extract the data of interest.”

Flexibility: AI does not have general scope. Instead, it is focused on particular use cases. This is known a “weak AI.”

This is why it is important to see if the application is built to handle your particular vertical or situation. 

“Take search, for instance, where AI can be used to re-rank results and improve relevance,” said Ciro Greco, who is the Vice President of Artificial Intelligence at  Coveo. “When applied to ecommerce, search is searching on semi-structured records, such as products with little text available and we can count on reasonable amounts of behavioral data produced by users who browse the website. A strategy based on user behavior can be very effective, because we can count on having enough data to learn from.”

Yet AI-search for customer service use cases is often much different. It’s often about finding technical documents. “There is plenty of unstructured text, such as Knowledge Articles, and fewer behavioral data points, because customer service websites usually are less visited than e-commerce platforms,” said Greco. “So in that case, a strategy based on NLP for topic modeling will probably be more effective, because we need to maximize the gain from the information we have, in this case free text.”

Ease-of-use: This is absolutely critical. The user of AI is often a non-technical person. If the application is complex, there could easily be little adoption.

Ethical AI: Even if the application is accurate, there could be risks. The data may have inherent biases, which could skew the results. This is why you should get an explanation of the data and how it is used. 

“What many forget when evaluating an AI solution is the potential damage or risk it could pose to your organization,” said Michael Mazur, who is the founder and CEO of AI Clearing. “What if your organization is sued for deploying this solution?”

Costs: “If you’re the first customer in a specific industry for an AI vendor, then you’re a very valuable customer to have and you can use that as negotiating leverage for a beneficial contract,” said Brian Jackson, who is an analyst and research director at Info-Tech Research Group.

Xometry IPO: Looking To Be The Airbnb Of On-Demand Manufacturing

It was a busy week for IPOs, as 18 companies issued shares. And one of the standout offerings was from Xometry, which operates an online marketplace for on-demand manufacturing.

The IPO was priced at $44, which was above the $38-to-$42 price range, and the shares soared nearly 100% on the first day of trading (the current market value is close to $3 billion). T. Rowe Price and Capital World Investors purchased $70 million of the shares in the offering.

The CEO and cofounder of Xometry is Randy Altschuler, who is a serial entrepreneur. He launched two other startups that were sold to pubic companies. 

As for Xometry, he teamed up with Laurence Zuriff (he is the current Chief Strategy Officer). Prior to this, he was the managing partner at Granite Capital International Group.

Both Altschuler and Zuriff were intrigued by the custom manufacturing industry. But they did not immediately create a company. Instead, they spent months researching the market by talking to many small manufacturers. “We saw certain themes emerge,” said Altschuler.

For example, smaller manufacturers were usually dependent on larger customers that were local, which posed considerable risk. They also spent much time responding to requests for custom parts that did not turn into business. 

The buyers of custom manufacturing parts also had challenges. It was difficult to find the best vendors and to come up with the right pricing. 

To solve these problems, Altschuler and Zuriff saw the need for building a two-sided marketplace. But it took some time to get to critical mass. While sellers were interested, there was skepticism from the buyers. 

Then there was the issue of the pricing for the marketplace. Custom jobs do not have SKUs. Rather, each one is unique.

This meant Xometry needed to build an automation system. One approach was to use brute-force and go through the possibilities. “The problem is that it would take too long to build such a system,” said Altschuler. “It would also be difficult and time-consuming to maintain it.”

The next approach then? It was to leverage AI (Artificial Intelligence). Xometry created a proprietary engine—backed with patents—to provide instant quoting based on factors like volume, material, location and the manufacturing process.

“What really got us excited about Xometry was that they were using incredible technology,” said Daniel Docter, who is a Managing Director at Dell Technologies Capital and an investor in the company. “Over time their AI algorithms got better, their modeling became more accurate, and their breadth of capabilities grew.”

The result is that the company has been able to efficiently process transactions for more than 6 million parts since inception. Currently there are over 43,000 buyers and 5,000 sellers on the platform (the customers include roughly 30% of the Fortune 500).

In terms of growth, it has been accelerating. From 2018 to 2020, the compound annual growth rate was 92%, with revenues hitting $141.4 million.

However, without the AI, none of this would have been possible. The technology has been strategic to Xometry .

“The AI learns and predicts how to make a part, how much it should cost, how long it should take, how much material would be needed, how would it likely yield, and so on,” said Doctor. “Xometry literally transforms an old, manual, grossly human limited process, to an automated, much more accurate and predictable, and hugely more efficient AI-driven process. Manufacturers win; customers win; and frankly the world wins as we collectively can do so much more with so much less waste.”

SentinelOne IPO: How The Company Is Riding The AI Wave

SentinelOne, which develops AI-powered software for cybersecurity, launched its IPO today. There was certainly substantial demand from investors. The initial price range was $26-to-$29 but this was lifted to $31-to-$32. The offering was then priced at $35 and the amount raised came to about $1.2 billion. Tiger Global, Insight Venture Partners, Third Point Ventures, and Sequoia Capital also participated in a $50 million concurrent private placement for the stock.

The CEO and cofounder of SentinelOne is Tomer Weingarten.  Before launching the company in 2013, he had helped to create several other tech startups. His background was mostly in analytics.

He would team up with Almog Cohen, who was a security expert at Check Point Software Technologies. Cohen and Weingarten were actually childhood friends and went to the same college.

As for SentinelOne, the vision was to build a next-generation cybersecurity platform that leveraged AI. But interestingly enough, the timing was too early. “In the first few years, it was an absolute battle to get the trust of customers,” said Weingarten.

Yet things started to change as the cybersecurity threats became more frequent and dangerous. The reality was that traditional systems—such as those based on human-powered signatures–were failing even more. It was “akin to bringing a knife to a gunfight,” according to the SentinelOne S-1 filing.

While building the AI system, Weingarten learned some important lessons. First of all, success would not involve building better algorithms. The reason? They tend to be similar, standardized and open source. 

Next, success with AI would not be about having huge amounts of data either. The focus instead should be on having the right data that produces signals that can be modelled. “We look at it as a contextual narrative, such as like telling a story,” said Weingarten. “We even received a patent on this approach.” 

Building the platform has required using the latest in data systems to process petabytes of data in real-time. Every second counts when it comes to fending off cyberattacks. “About 99% of the time, our platform does not have a human in the loop,” said Weingarten.

The SentinelOne system is flexible as well. For example, it can be deployed on environments like Windows, macOS, Linux, and Kubernetes.

The Growth Engine

Consider that none of the company’s customers were impacted by the SolarWinds Sunburst cyberattack. This was definitely a major validation of the AI approach.

And yes, the growth has been standout for SentinelOne. During the latest quarter, revenues soared by 108% to $37.4 million. There are currently more than 4,700 customers and a majority of them are large enterprises. 

Some of the other metrics include:

  • The dollar-based gross retention rate: 97%.
  • The dollar-based net retention rate: 124%.
  • Annual recurring revenue: $164 million.
  • Customer satisfaction: 97%.

Note that the SMB (small-and-medium size business) category has shown even more growth. A key has been the leveraging of MSPs (Managed Service Providers).


This was the first public offering for Weingarten. “It was a lot of hard work,” he said.  “But we thought that an IPO was critical. I think it is about becoming a more mature company.”

Being public also helps with the trust of customers. After all, there are stringent disclosure and audit requirements. In fact, some larger enterprise companies will not even purchase cybersecurity software from private companies.

Now it’s true that SentinelOne faces intense competition. Just some of the key rivals include CrowdStrike and Palo Alto Networks.

Yet the market is massive. Based on the analysis from IDC, the spending is expected to reach $40.2 billion by 2024, which represents a compound annual growth rate of nearly 12%.

Biden’s AI Initiative: Will It Work?

The Biden administration has recently set into action its initiative on AI (Artificial Intelligence). This is part of legislation that was passed last year and included a budget of $250 million (for a period of five years). The goals are to provide easier access to the troves of government data as well as provide for advanced systems to create AI models. 

No doubt, this effort is a clear sign of the strategic importance of the technology. It is also a recognition that the U.S. does not want to fall behind other nations, especially China. 

The AI task force has 12 distinguished members who are from government, private industry and academia. This diversity should help provide for a smarter approach.

But the focus on data will also be critical.  “In areas of social importance such as housing, healthcare, education or other social determinants, the government is the only central organizer of data,” said Dr. Trishan Panch, who is the co-founder of Wellframe. “As such, if AI is going to deliver gains in these areas, the government has to be involved.”

Yet there will certainly be challenges. Let’s face it, the U.S. government often moves slowly and is burdened with various levels of local, state and federal authorities. 

“To achieve the initiative’s vision, government entities will need to go beyond sharing best practices and figure out how to share more data across departments,” said Justin Borgman, who is the CEO of Starburst. “For instance, expanding open data initiatives which today are largely siloed by departments, would greatly improve access to data. That would give Artificial Intelligence systems more fuel to do their jobs.”

If anything, there will be a need for a different mindset from the government. And this could be a heavy lift. “Based on my experience in the public sector, the major challenge for the government is addressing the ‘Missing Middle,’” said Jon Knisley, who is the Principal of Automation and Process Excellence at FortressIQ.   “There are a number of very advanced programs on one end, and then there are a lot of emerging programs on the other end. The greatest opportunity lies in closing that gap and driving more adoption. To be successful, there should be a focus as much as possible on applied AI.”

But the government initiative can do something that has been difficult for the private sector to achieve—that is, to help reskill the workforce for AI. This is perhaps one of the biggest challenges for the U.S. 

“The question is: How do we create a large AI data science force that is integrated across every industry and department in the US?,” said Judy Lenane, who is the Chief Medical Officer at iRhythm. “To start, we’ll need to begin AI curriculum early and encourage its growth in order to build a comprehensive workforce. This will be especially critical for industries that are currently behind in technological adoption, such as construction and infrastructure, but it also needs to be accessible.”

In the meantime, the Biden AI effort will need to deal with the complex issues of privacy and ethics. 

“Presently there is significant resistance on this subject given that most consumers feel that their privacy has been compromised,” said Alice Jacobs, who is the CEO of convrg,ai. “This is the result of a lack of transparency around managing consents and proper safeguards to ensure that data is secure. We will only be able to be successful if we can manage consents in a way where the consumer feels in control of their data. Transparent unified consent management will be the path forward to alleviate resistance around data access and can provide the US a competitive advantage in this data and AI arms race.”

Can AI Solve Your Hiring Problems?

In April, the number of job openings hit a record 9.3 million, according to data from the Labor Department. With the pandemic fading away, there has been a scramble to hire new employees and this has become a major challenge for companies.

So then can AI (Artificial Intelligence) help out? Well, it definitely can. The irony is that many companies are using the technology—and don’t even realize it! The reason is that AI is built into the top online job sites. 

“For example, when you type in a search for a job title, say with the phrase ‘job manager,’ the LinkedIn engine will not only look for the title itself, but also people with relevant skills like time management, team coordination, risk assessment and so on,” said Sakshi Jain, who is the Engineering Manager on LinkedIn’s Responsible AI team. “This means that a recruiter gets more results than just the people who already have the exact title or role.”

One of the key powers of AI is that it can detect complex patterns in huge data sets.  In a way, this can simulate the capabilities of a recruiter. And this is definitely important when it comes to finding passive job candidates. 

“In a study we recently published, 74% of talent leaders told us they’ve increased outreach to passive candidates in the past year,” said Hari Kolam, who is the CEO of Findem. “AI greatly speeds up the passive recruiting process–one where it can take upward of ten hours to fill a single role. It can index and surface information on people from hundreds of sources as passive candidates typically aren’t on job or career sites, and many people tend to only include piecemeal information on their LinkedIn and other profiles.”

AI can also help with personalization. This can be a good way to create a good first impression with candidates.

“Currently, there are many hiring workflows that are incredibly inefficient, including the scheduling of interviews and follow-up emails for candidates,” said Vivek Ravisankar, who is the CEO of HackerRank. “With the use of automated scheduling and email follow-ups, AI can help free up valuable time and solve the major pain point of extensive back-and-forth coordination with candidates and interviewers.”

Yet AI is not without its risks. After all, there is inherent bias in datasets and this can result in outcomes that are unfair and discriminatory. 

“AI-driven HR software that is using years old data on previous hires to determine ideal candidates for job openings is a perfect example of where algorithms can go wrong,” said Ingrid Burton, who is the Chief Marketing Officer at Quantcast. “This is especially true in roles that have historically been dominated by men, such as software engineers, which would risk the hiring algorithm to arbitrarily exclude most women and minorities from advancing during the hiring process.”

To guard against this, there must be good governance as well as explainability of the AI models. There also needs to be people in the loop for critical parts of the process. 

“It is never acceptable to set up an AI process and simply leave it to run,” said Ian Cook, who is the Vice President of People Analytics at Visier. “While there is no need to inspect every transaction or process run by the AI, there is a need to constantly review the performance of the AI steps to ensure that the outputs are in line with expectations. Validation, updating and retraining are constant requirements of running any AI process.”’s Tom Siebel: How To Scale AI, which is a top provider of enterprise AI software and services, is a newly public company. It pulled off its deal in December and issued 15.1 million shares at $42 each. On the first day of trading, the shares spiked 120%.

And this would not be the end of the gains. Within a couple months, the stock price would hit an all-time high of $183.

But as the markets started to cool off, so did the shares of Consider that the stock price is now at $64.

Despite this, the future does look bright for the company. “The total addressable market is huge,” said Tom Siebel, who is the CEO and founder of “It’s a third of a trillion dollars.”

Keep in mind that Siebel is a veteran of the enterprise software world. In the early 1980s, he worked as an executive at Oracle and helped make the company the dominant player in relational databases. Then in 1993, he started Siebel Systems and pioneered the CRM (Customer Relationship Management) category.

As for, he launched this company in 2009. Siebel was early in recognizing that AI would be a megatrend. 

But he also crafted a solid approach to building the platform. “We were novel in using a model-driven architecture to enable organizations to rapidly design, develop, provision and operate enterprise AI applications at scale,” said Siebel. “We spent about a billion dollars inventing this in the last decade and it is our secret sauce.”

This was in contrast to using traditional techniques, such as with structured programming, that involve a mishmash of open source and proprietary solutions. However, this usually means too much complexity to effectively scale. 

Even some of the world’s top companies have suffered major blunders and failures with AI. IBM’s Watson, for example, has fallen well short of expectations. Then there is GE, which has spent billions on AI and has seen little return. 

The Future Of Enterprise AI

The platform can handle applications for global enterprises as well as small businesses. And to get a sense of its power, it currently manages over 4.8 million concurrent production AI models and processes more than 1.5 billion AI predictions per day. 

Now, another key factor for the success of is that the company takes a partnership approach with customers. This is essential for AI since it is important to leverage vertical-specific data and insights. 

A case study for this is Shell. “The company is reinventing itself as the fifth largest in the world,” said Siebel. “They want to get to a zero net carbon footprint by 2050, which is no mean trick, right? This is about applying AI to the entire value chain, about delivering cleaner, safer, lower cost, more reliable energy. This is maybe a $4 billion a year economic benefit.”

Granted, the temptation for companies is to build their own systems. But this is really the wrong approach. “Believe it or not, people tried to build their own relational database systems during the 1980s and I don’t think anybody succeeded,” said Siebel. “We are seeing it again with AI. Companies will try it once, twice, three times. Then they’ll wind up firing the CIO and buy the technology from a professional.”

Ultimately, Siebel thinks that AI will be similar to CRM or ERP. In other words, it will be a technology that’s a necessity for a large number of businesses. “Companies that do not adopt AI will no longer exist,” said Siebel. 

AI (Artificial Intelligence): How Non-Tech Firms Can Benefit

Even though AI continues to thrive and grow, there remain challenges to use the technology. Just some include finding data scientists, determining the right problems to focus on, getting quality data and scaling the models.

No doubt, these problems are even worse for non-tech companies. They generally do not have the expertise or sufficient resources to make AI a success.

“Research shows non-tech companies in particular have struggled to take their AI programs beyond the proof of concept and pilot phases–with just 21% of retail, 17% of automotive, 6% of manufacturing, and 3% of energy companies successfully scaling their AI use cases,” said Jerry Kurtz, who is the Executive Vice President of Insights and Data at Capgemini North America.

But despite all this, there are still a myriad of companies that are beating the odds. And they are becoming much more competitive. “There are many opportunities for non-tech companies to leverage AI to improve efficiency and provide a better customer and employee experience,” said Margaret Lee, who is the Senior Vice President and General Manager of Digital Service Operations Management at BMC.

So what are some of the non-tech companies that have been able to move-the-needle with their AI efforts? Here’s a look at two and the lessons learned.

John Deere: Keep in mind that the company has a long history of innovation, going back to the invention of the steel plow in 1837. The result is that John Deere is a world leader, with a market capitalization of $120 billion and annual sales of over $35 billion.   

The company’s AI efforts began with machine vision because of its advantages with GPS connectivity in its equipment and rich datasets. For example, during the spring, the peak data ingestion was 425MB per second or about 50 million sensor measurements per second. 

“Our projects have resulted in products in the market, or soon to be in the market, that reduce chemical inputs but sensing weeds from non-weeds and selectively apply chemical to the weeds only,” said Jahmy Hindman, who is the Chief Technology Officer of John Deere. “In addition, this work has led to vision-based automated control systems in combine harvesters that optimize the processing settings of the ‘factories on wheels’ to minimize the grain lost during harvest.”

A key lesson for the company has been the importance of keeping customer needs and wants in mind. “We work tirelessly to help our customers advance their business and feed the world,” said Hindman.

Levi Strauss: An early project for this company came in response to the Covid-19 lockdowns in Europe. Levi Stratus looked for ways to manage the inventory pile-up. To this end, the company gathered unique datasets on dynamic price elasticities and then applied AI to it. 

It was a small project but was able to scale quickly. What started as a test in 11 stores in Germany in May 2020 grew to 17 countries in Europe by October.  The system was also used for the 11/11 Singles’ Day in China. 

“I have three pieces of advice from this experience,” said Louis DiCesari, who is the Global Head of Data, analytics, and AI at Levi Strauss. “First, choose real, commercial problems that are aligned to your company’s strategic priorities, and chunk them into actionable steps. Second, don’t get too hung up on having perfect data or technology, or using the latest algorithms. Instead, embrace agile, deliver minimum viable products, continuously measure the impact, and continue to iterate and add new features. And, of course, set a vision, communicate the progress toward that vision throughout the organization, and invite feedback.”

DiCesari attributes AI to the company’s ability to accelerate innovation and move faster than ever before. “In 2021, we aim to deliver more value and support all countries and functions of the enterprise, infuse data and AI throughout the business, enable new ways of working and continue streamlining processes, while also digitizing assets,” he said.

The Edge: What Does It Mean For AI (Artificial Ingelligence)?

The edge is an end point where data is generated through some type of interface, device or sensor. Keep in mind that the technology is nothing new. But in light of the rapid innovations in a myriad of categories, the edge has become a major growth business. 

“The edge brings the intelligence as close as possible to the data source and the point of action,” said Teresa Tung, who is the Managing Director at Accenture Labs. “This is important because while centralized cloud computing makes it easier and cheaper to process data at scale, there are times when it doesn’t make sense to send data off to the cloud for processing.”

This is definitely critical for AI. The fact is that consumers and businesses want super-fast performance with their applications. 

“Currently AI training produces vast volumes of data that are almost exclusively implemented and stored in the cloud,” said Flavio Bonomi, who is the board advisor to Lynx Software. “But by placing compute at the edge, this allows for looking at patterns locally. We believe this can evolve the training models to become simpler and more effective.”

The edge may even allow for improved privacy with AI models. “Having federated learning means that no end-user data is centralized or communicated between nodes,” said Sean Leach, who is the Chief Product Architect at Fastly.

What Can Be Done At The Edge

The most notable use case for the edge and AI is the self-driving car. The complexities are mind boggling, which is why the development of this technology has taken so long.

But of course, there are many other use cases that span a myriad of industries. Just look at manufacturing.  “In monitoring manufacturing processes where seconds or minutes could mean millions of dollars in losses, for example, machine learning models embedded in sensors and devices where the data is being collected enables operators to preemptively mitigate serious production issues and optimize performance,” said Santiago Giraldo, who is the Senior Product Marketing Manager of Machine Learning at Cloudera.

Here are some other examples:

  • Chris Bergey, the Senior Vice President and General Manager of Infrastructure Line of Business at Arm: “AI and the edge can explore the impacts of urbanization and climate change with software-defined sensor networks, pinpoint the origins of power outages in smart grids with data provenance, or enhance public safety initiatives through data streaming.”
  • Adam Burns, the Vice President of IoT and the Director of Edge Inference Products at Intel: “CORaiL, which was a project with Accenture and the Sulubaaï Environmental Foundation, can analyze coral reef resiliency using smart cameras and video analytics powered by Intel Movidius VPUs, Intel FPGAs and CPUs, and the OpenVINO toolkit.”
  • Jason Shepherd, the Vice President of Ecosystems at ZEDEDA: “TinyML will enable AI in more appliances, connected products, healthcare wearables, etc., for fixed functions triggered locally by simple voice and gesture commands, common sounds (a baby crying, water running, a gunshot), location and orientation, environmental conditions, vital signs, and so on.”
  • Michael Berthold, the CEO and cofounder at KNIME: “In the future, we will also see models that update themselves and potentially recruit new data points on purpose for retraining.”
  • Ari Weil, who is the Global Vice President of Product and Industry Marketing at Akamai: “Consider medical devices like pacemakers or heart rate monitors in hospitals. If they signal distress or some condition that requires immediate attention, AI processing on or near the device will mean the difference between life and death.”

But successfully bringing AI to the edge will face challenges and likely take years to get to critical mass.  “The edge has relatively lower resource capabilities in comparison to data centers, and edge deployments will require lightweight solutions focused on security and supporting low latency applications,” said Brons Larson, who is a PhD and the AI Strategy Lead at Dell Technologies.

There will also need to be heavy investments in infrastructure and the retooling of existing technologies. “For NetApp, this is a large opportunity but one that we have to re-invent our storage to support,” said Ross Ackerman, who is the Head Of Customer Experience and Active IQ Data Science at NetApp. “A lot of the typical ONTAP value prop is lost at the edge because clones and snapshots have less value. The data at the edge is mostly ephemeral, needing only a short time to be used in making a recommendation.”

Then there are the cybersecurity risks. In fact, they could become more dangerous then typical threats because of the impact on the physical world. 

“As the edge is being used with applications and workflows, there is not always consistent security in place to provide centralized visibility,” said Derek Manky, who is the Chief of Security Insights and Global Threat Alliances at Fortinet’s FortiGuard Labs. “Centralized visibility and unified controls are sometimes being sacrificed in favor of performance and agility.”

Given the issues with the edge and AI, there needs to be a focus on building quality systems but also rethinking conventional approaches. Here are some recommendations:

  • Prasad Alluri, the Vice President of Corporate Strategy at Micron: “The increase in AI also means that its increasingly important that edge computing is near 5G base stations. So soon, in every base station, every tower might have compute and storage nodes in it.”
  • Debu Chatterjee, the Senior Director of AI Platform Engineering at ServiceNow: “There will need to be newer chips with tensor capabilities seen in GPUs or their alternative, or specialized with specific inference models burnt into FPGAs. A hardware/software combo will be required to provide a zero-trust security model at the edge.”
  • Abhinav Joshi, the Global Product Marketing Leader at OpenShift Kubernetes Platform at Red Hat: “Many of these challenges can be successfully addressed at the start by approaching the project with a focus on an end-to-end solution architecture built on the foundation of containers, Kubernetes, and DevOps best practices.”

Although, when it comes to AI and the edge, the best strategy is probably to start with the low-hanging fruit. This should help avoid failed projects.

“Enterprises should begin by applying AI to smaller, non-mission critical applications,” said Bob Friday, who is the Chief Technology Officer at Mist Systems, which is a Juniper Networks company. “By paying close attention to details such as finding the right edge location and operational cloud stack, it can make operations easier to manage.”

But regardless of the approach, the future does look promising for the edge.  And AI efforts really need to consider the potential use cases to get its full value.

Snowflake IPO: What You Need To Know

With the equity markets surging and interest rates at historically low levels, the environment is ideal for IPOs. But for Snowflake, which has recently filed for an offering, it would likely do well in just about any market environment. This tech startup is growing like a weed and the market opportunity is enormous.

Founded in 2012, Snowflake pioneered the category for cloud-native data warehouses. The founders actually spent two years developing the software.

And yes, the timing proved to be spot-on. The market was ripe for disruption as traditional data warehouses have a myriad of disadvantages. Just some include: the inability to handle unstructured data and huge workloads, high costs, complex interfaces, problems with consistency and integrity of data, and issues with data sharing.   

Note that the founders—Thierry Cruanes, Benoit Dageville, and Marcin Zukowski—were veterans of the traditional data warehouse market. They had worked at companies like Oracle, IBM and Google (by the way, the name “Snowflake” was chosen because the founders like to ski!) In other words, the founders had a strong understanding of the weaknesses of legacy systems—but also had the creativity to build a much better alternative.

“My company uses Snowflake and also competes with it in some cases,” said Sam Underwood, who is the VP of Business Strategy with Futurety. “Snowflake is growing rapidly, and justifiably so, because it’s filling a gaping hole in the market—namely, a huge need to unify data sources to form a single source of truth across an organization. There are many, many tools that already do this—Google BigQuery among others—however, Snowflake has combined the technical effectiveness with the UI simplicity to really excel among both technical users and high-level decision makers who may not want or need as much granular detail.”

So how fast is Snowflake growing? During the first six months of this year, revenues spiked from $104 million to $242 million on a year-over-year basis. While there continues to be significant net losses, the company has still been able to greatly improve gross margins. 

Consider that a key technology decision for Snowflake was to separate compute from storage. “This offers great performance to customers without the high cost, so they get the best of both worlds,” said Venkat Venkataramani, who is the co-founder and CEO of Rockset. “This was phenomenally compelling at the time and years ahead of even the likes of Amazon with Redshift and Google.”

But of course, Snowflake is more than just about whiz-bang technology. The company has also assembled an experienced executive team, led by CEO Frank Slootman. Prior to joining, he was at the helm of ServiceNow, which he took from $100 million in revenues to $1.4 billion. The current market cap of the company is $93 billion. 

True, Snowflake does have customer concentration, with Capital One accounting for roughly 11% of overall revenues. But then again, this does show the strategic importance of the technology. 

“This IPO underscores a significant change in thinking about the increasing importance of the database market,” said Raj Verma, who is the Co-CEO of MemSQL. “Data has never been more important than it is right now. In the last 25 years, only one company in this sector other than Snowflake went public. And I’m sure we’ll see a couple more companies go out in the new few years as well. There was an iron grip on the database market for more than two decades, with IBM, Oracle and SAP HANA. Now we are seeing a changing of the guard, which gives customers the option of deciding what is best for their business. I can tell you that the technology of yesterday will not solve the data challenges of tomorrow, and this IPO brings newer technology solutions to the forefront.”

Quantum Computing: What Does It Mean For AI (Artificial Intelligence)?

While quantum computing is still in the early phases, there have already been many innovations and breakthroughs. Companies like IBM, Microsoft, Google and Honeywell have been investing aggressively in the technology. 

So then what is quantum computing? Well, it is similar to traditional computing, which relies on bits—that is, the 0’s and 1’s to encode information. But quantum computing as its own version of this: the quantum bit or qubit. This is where the information can have multiple states at the same time. And the reason for this is the impact of the effects of quantum mechanics, like superposition and entanglement. Yes, this is all about the spooky world of Schrodinger’s cat, which is both alive and dead at the same time!

“Quantum computing is a new kind of computing, using the same physical rules that atoms follow in order to manipulate information,” said Dr. Jay Gambetta, who is an IBM Fellow and vice president of IBM Quantum. “At this fundamental level, quantum computers execute quantum circuits—like a computer’s logical circuits, but now using the physical phenomena of superposition, entanglement, and interference to implement mathematical calculations out of the reach of even our most advanced supercomputers.”

One of the fertile areas for quantum computing is AI (Artificial Intelligence), which relies on processing huge amounts of complex datasets. There is also a need to evolve algorithms to allow for better learning, reasoning and understanding.

Then what are some of the things we may see with quantum computing and AI? Let’s take a look:

Christopher Savoie, who is the CEO and founder of Zapata Computing:

Generative models are those models that don’t just limit themselves to answering a question, but that actually generate output such as an image, music, video, etc. As an example, imagine you have a lot of pictures of the side of a face, but not a lot of pictures of the front of a face. If you want security detection capabilities to be able to recognize dual facial recognition on the front side of a face, you can actually use these generative models very accurately to create more samples of frontal views of a face. Inserting quantum processing units into the classical framework has the potential to boost the quality of the images generated. And how does this help us with classical machine learning? Well, traditional machine learning algorithms are as good as the data you feed them. If you try to train a classical face detection model with a small dataset of faces, this model won’t be very good. However, you can use quantum-enhanced generative models to enlarge this dataset with more images (both in terms of quantity and variety), which can significantly improve the detection model. This isn’t limited to generating faces, you can also generate fake molecules, cancer cells, or MRI scans, which are very similar to the real thing. This allows us to train better machine learning models, which can then apply to real data and real-world problems.

Ilyas Khan, who is the CEO of Cambridge Quantum Computing:

For the first time, a Natural Language Processing (NLP) algorithm is “meaning aware” and has been executed on a quantum computer. When we refer to meaning aware we mean that computers can actually understand whole sentences and not just individual words and that the awareness can be expanded to whole phrases and ultimately real time speech without requiring stochastic guesswork that is the state of the art today and which is computationally so expensive. Full scale implementation is dependent on quantum computers becoming much larger than is currently the case. This development of research in NLP is a prime example of the fact that as realistic quantum computers become available, more use cases will also become apparent. Of course, this has been proven to be the case in the past 30 years on classical computers as a precedent.

Dr. Itamar Sivan, who is the CEO and co-founder of Quantum Machines:

Roughly speaking, AI and ML are good ways to ask a computer to provide an answer to a problem based on some past experience. It might be challenging to tell a computer what a cat is, for instance. Still, if you show a neural network enough images of cats and tell it they are cats, then the computer will be able to correctly identify other cats that it did not see before. It appears that some of the most prominent and widely used AI and ML algorithms can be sped-up significantly if run on quantum computers. For some algorithms we are even anticipate exponential speed-ups, which clearly does not mean performing a task faster, but rather turning a previously impossible task and making it possible, or even easy. While the potential is undoubtedly immense, this still remains to be proven and realized with hardware.

Tony Uttley, who is the President of Honeywell Quantum Solutions:

One of the areas being looked at currently is in the area of artificial intelligence within financial trading. Quantum physics is probabilistic, meaning the outcomes constitute a predicted distribution. In certain classes of problems, where outcomes are governed by unintuitive and surprising relationships among the different input factors, quantum computers have the potential to better predict that distribution thereby leading to a more correct answer. Dr. Hayes states:  “The basic idea is that there are problems that require an AI to generate new data that it hasn’t seen before in order to make a decision. Solving this problem may require coming up with an underlying model for the probability distribution in question that it could use in new situations.”

Daniel Newman, who is the Principal Analyst and Founding Partner at Futurum Research:

As it pertains to AI/ML, I think what I’m most encouraged by is the potential for classical and quantum to work together leveraging the elastic nature of the cloud and the powerful, specific problem-solving capabilities of quantum computing. I get the sense that a lot of people are looking at quantum versus classical computing, but in reality, it will be the two working together harmoniously to solve challenging and complex problems. Both have strengths and the development now is for quantum computing to function as part of the solution. Over time, both computing formats will continue to advance, but the ability to accelerate workloads on traditional GPUs and ASICs while also leveraging the power of quantum computing is a recipe for faster, more robust results, which is what the market should be eager to see as quantum computing becomes more widely accessible.

For me, I see a few applications for quantum computing in the immediate future that will gain popularity, but of course there will be many more. Financial Services and Healthcare are immediate applications where Quantum Computing can take advantage of speed and specificity to help tackle complexities. Fraud detection and drug compound identifications have been touted as some of the most exciting use cases. Given the current state of cybercrime and the attention to healthcare in the wake of the pandemic, this makes a lot of sense.