Machine learning observability built for practitioners: Our investment in Arize

We’re still in the early innings of AI adoption, yet we’ve already seen it transform industries. Companies like Netflix, Spotify, and Uber have scaled internal teams from a handful of data scientists and machine learning (ML) engineers to hundreds. These teams, and the models they built to inform business and product decision-making, have shaped how consumers watch television, listen to music, and hail rides.

As AI use cases abound, ML teams face growing challenges around how to build, deploy, and maintain their models. Practitioners demand the flexibility to optimize each component of a model and, like software development, use specific tools to address the various phases of the model development lifecycle. In an attempt to bridge the development gap in ML, a new framework called MLOps has emerged. MLOps is derived from core DevOps principles and represents one of the fastest growing markets in technology today. 

To date, MLOps has largely centered on data preparation, model training, and model deployment; however, building and deploying models is only the beginning of the journey. What happens once a model is running in production? The reality is most companies still lack the necessary tools to scalably monitor and understand their live models. Moreover, as these models become more complex, troubleshooting issues gets harder and both upstream and downstream problems compound. 

In order for AI to achieve long-term sustainability, companies must improve model transparency, understanding, and performance. Enter Arize, a ML observability platform built by practitioners to help unpack the proverbial AI black box and optimize the performance of models in production. 

Shaping the future of AI infrastructure

As the various components of the model lifecycle standardize, many companies have started to orient away from expensive in-house builds and less flexible integrated platforms. Instead, practitioners are opting for best-of-breed MLOps solutions from focused vendors like Arize to gain additional control over their modeling workflow.

Arize sets itself apart as one of the few platforms that sits at the center of production ML. Within MLOps, there has been a flood of investment in tools addressing data preparation all the way through to the model deployment stage, but few tackle the reality that all live models face over time: degradation. Without real-time monitoring and observability, ML teams spend countless hours pouring over anomalies and trying to understand problems in the data, software, and / or model itself. Arize’s observability platform seamlessly plugs into any MLOps stack and provides a scalable solution to monitor the performance of models, explain what the models are trying to do, and diagnose data and drift issues without going back to square one.

In speaking to Arize’s customers, which include many of the world’s most sophisticated ML teams, it’s clear observability is seen as a core pillar of AI infrastructure and represents a natural progression in how they think about model lifecycle management. There’s a reason adjacent observability solutions like Datadog and Monte Carlo exist in other areas of IT, and we believe ML will be no different. 

Built by practitioners

Arize’s founders, Jason Lopatecki and Aparna Dhinakaran, first met at TubeMogul, where Jason was a founder who helped build out the company’s ML team and Aparna was a data scientist. Jason would eventually guide TubeMogul through a successful IPO and sale to Adobe while Aparna went to work for Uber as part of its famed Michelangelo team. 

Jason and Aparna stand out in a MLOps space where many founders hail from academia. Both draw from deep practitioner roots and have experienced firsthand the heartache of spending months building and training models, deploying them to production, and having no insight into how the models actually performed once deployed. Independently, they came to the conclusion that something was fundamentally missing in the MLOps toolchain. Together, they are now focused on bringing transparency, understanding, and performance to production ML through Arize’s dedicated observability platform. 

Our partnership with Arize

We’re thrilled to announce that TCV has led Arize’s $38M Series B alongside our friends at Battery Ventures, Foundation Capital, and Swift Ventures. 
At TCV, we gravitate towards founders that are culture and product obsessed. Jason and Aparna blend multi-stage company-building experience with firsthand knowledge of a real customer pain point. We’re incredibly excited to partner with the Arize team on their mission to make AI work and work for the people. If you are interested in joining Arize for the journey ahead, please visit their website to learn more about current career opportunities.


TCV invests in Evisort to deliver scalable, AI-powered contract management

Contracts are at the heart of business, enshrining a company’s rights and obligations across areas ranging from sales transactions and supplier relationships to employment agreements and beyond. Resulting from this centrality, rising contract volumes and legal complexity have made contract management unmanageable without leveraging technology.

Evisort delivers end-to-end contract intelligence software that turns contracts into data. Customers use a simple, intuitive interface to extract critical context from contracts, integrate that data into other enterprise systems, and automate a wide range of legal and operational workflows – themselves codified in contract data. Evisort’s platform is powered by award-winning AI that is purpose-built for contracts and trained on over 10 million documents, thereby driving a differentiated customer experience and rapid, tangible ROI.

We are thrilled to announce TCV’s Series C investment in Evisort. We believe that contracts have been both an under-managed source of risk and under-explored source of value for companies, and that Evisort’s AI-powered Contract Intelligence Platform solves increasingly important pain points for businesses of all sizes, ranging from the Fortune 500 to mid-sized companies alike.

Evisort was founded in 2016 by lawyers and technologists who saw the need for automation in contract management.  The platform started as an intelligent analytics engine that extracts clauses and metadata to index contracts and their contents, making them easily searchable and manageable without manual data entry. Evisort’s AI further contextualizes the contract, indicating what type of contract it is, identifying counter-parties, flagging auto-renewal dates, and more.

More recently, Evisort has been adding workflow capabilities – relevant for coordinating contracting processes and operational workflows across the business. Evisort’s end-to-end approach ensures that all contract data is located in one repository, minimizing security risks, reducing the number of required integrations, and allowing the system to apply learnings from previous contracts to new ones.

More than any other contract management software business we’ve evaluated, Evisort’s AI platform supports a wider range of teams, industries, and use cases. Sales teams use Evisort to drive sales and renewals by reducing contracting friction and speeding time to agreement and revenue recognition. Legal departments use Evisort to drive compliance, quickly find and report on critical information, and act as a single source of truth. Procurement and sourcing organizations rely on Evisort to accelerate purchases, negotiate stronger agreements, and manage supplier risks more effectively. In all cases, Evisort drives efficiency by reducing reliance on manual legal review – a major bottleneck in many contracting processes.

Transforming the future of contract management

Evisort’s Contract Intelligence Platform has three main capabilities:

AI-Powered Contract Analytics and Insights: Evisort extracts data from contracts, produces critical insights, and reports on those insights in an easy-to-use dashboard, so that users can focus on higher value tasks. This contract intelligence is then used to generate workflows across the organization. Evisort is focused on delivering the intelligence layer between core operating systems such as customer relationship management and enterprise resource planning platforms.

Intelligent Contract Lifecycle Management: Evisort provides contract request intake, contract drafting, approvals, version control, and repository (storage, search, reporting) features. Evisort’s platform creates a source of truth so teams can centralize knowledge, collaborate easily, and simplify contract administration.

Central Contract Repository and Integrations: Evisort’s no-code platform lets legal, sales, and procurement teams self-serve, taking the burden off of IT teams and providing immediate configurability. Evisort easily integrates into existing systems to minimize the need for data migration and accelerates deployment because employees can work from the systems they already use.

Why now: A big market waiting for the right end-to-end product

At TCV, we have invested extensively behind the digitization of the legal industry – having backed innovative legal technology industry leaders such as Clio, LegalZoom, and Avvo. As part of our work in this space, we have been closely following the evolution of the CLM market for nearly a decade. In that time, customers consistently indicated a desire to manage both new and existing contracts in the same place – in other words, a true end-to-end platform. Over the last several years, our conversations in the space increasingly indicated that Evisort’s founders Jerry, Jake, and Amine had built exactly that and Evisort’s platform was seeing accelerating adoption in a largely greenfield market.

Evisort customers – which include our portfolio companies such as Netflix – typically start with analytics use cases to understand existing contracts, and then add pre-signature workflow to more efficiently generate new contracts. From there, thanks in part to Evisort’s ease of use, usage often quickly expands to additional teams and stakeholders within their organization. For customers, the results are industry-leading time-to-value, implementation speeds, self-service analytics, and flexibility to apply contract-based insights to a wide range of business functions. For Evisort, a cohesive and forward-thinking strategy appears to have translated into an innovative and fast growing company in an exciting market.

Looking Forward

As we look to the future, we are incredibly excited about the tailwinds strengthening Evisort’s value proposition for its customers. Businesses of all sizes have more contracts and a greater need to manage them than ever before. The compliance and regulatory environment also continues to evolve, requiring businesses to maintain constant visibility into their contract corpus. And companies are increasingly leveraging the data embedded in contracts to drive business processes across sales, procurement, operations, and finance.

Given that robust backdrop, we are incredibly excited to work with Jerry, Jake, Amine and the rest of the Evisort team to maximize the opportunity for AI applications in contract management.

 


Shifting the R&D paradigm through AI/ML technology: Introducing BenchSci

TCV’s healthcare team has long been pursuing a thesis around the utilization of healthcare data, particularly for applications in the life sciences industry. Specifically, we believe that companies with proprietary technology that enables them to aggregate, curate, and contextualize healthcare data have a tremendous opportunity to layer on software applications and help address a myriad of downstream use cases for their customers. Our Series C investment in BenchSci, completed in partnership with our friends at iNovia Capital and F-Prime Capital, provides an illustration of this thesis in our portfolio – one of many, we hope, over the next few years. The Series C funding is intended to help BenchSci expand the company’s artificial intelligence and machine learning-enabled software platform into additional applications, rapidly scale headcount, and forward invest in future growth initiatives.

BenchSci was founded in 2015 by CEO Liran Belenzon, Chief Science Officer Tom Leung, Chief Data Officer Elvis Wianda, and co-founder David Chen who met one another through the University of Toronto’s Creative Destruction Lab. The company’s technology platform endeavors to increase productivity and efficiency in the preclinical research process for pharmaceutical and biopharmaceutical organizations. The life sciences industry spends an extraordinary amount on preclinical research as these efforts help develop a pharmaceutical or biopharmaceutical company’s core intellectual property. We estimate global preclinical expenditures at ~$80B annually, or ~40% of total research and development investment for life sciences firms, and scientists at pharmaceutical and biopharmaceutical companies perform tens of thousands of preclinical experiments per year. 

Despite this level of investment, preclinical research has long been plagued by inefficiencies. BenchSci’s customers estimate that approximately 80% or more of preclinical experiments performed yield no value to their overall research efforts; relatedly, it is extremely challenging to identify potentially wasteful or redundant experiments a priori. This process continues to be one of trial and error – more “art than science” – and limited technology tools exist to help scientists become more productive. Moreover, the data captured in the context of these efforts exist in disparate, siloed systems, thereby inhibiting information sharing and collaboration even within a life sciences company. Even when successful, a preclinical research process takes between six to seven years on average, thereby delaying time-to-market for life-saving medicines.

The problem described above is the one CEO Liran and his team are determined to solve via technology. The company’s mission is to deliver technology that helps scientists bring novel medicines to market 50% faster by 2025. To do so, BenchSci has built a comprehensive preclinical experiment-focused knowledge graph, encompassing data on over 40 million experiments.  Consistent with our framework outlined above, the company has built software and computer vision technology that automates the stitching together and curation of experimental, bioinformatic, and other data from numerous, disparate sources, including its customers’ own internal data. Further, the company’s team of PhD scientists works alongside BenchSci’s product and technology teams to contextualize BenchSci’s 100+ machine learning models and algorithms such that its knowledge graph and results make “scientific sense” to scientist end-users as they leverage the company’s technology platform.

BenchSci’s flagship application was launched in 2017, and leverages artificial intelligence to help scientists select the optimal antibodies and/or reagents to use in their experiments based on experiments previously performed that are relevant to the study in question. This saves scientists significant time and resources – customers indicated to us that they have saved tens of millions in hard costs alone by eliminating redundant/wasteful reagent purchases, not to mention the time savings (several weeks to months per project) and other efficiencies they’ve realized. The company is not stopping there, and we are particularly excited about what BenchSci is going to do next with its breakthrough technology that will shape the future of preclinical research, although we will leave it to Liran and his team to share more in the coming months.

BenchSci’s compelling value proposition, coupled with its reputation for relentless innovation and superb customer service and support, has engendered customer delight, and the company boasts a net promoter score of 80+. Its customers include 16 of the top 20 pharmaceutical companies (by revenue), in addition to over 4,500 leading research centers globally, and its platform is being used regularly by 50,000+ scientists. CEO Liran has scaled the organization to meet latent demand – BenchSci has grown its employee base more than 8x in the past three years, and expects to reach 400+ employees by the end of 2022. The company has been recognized as a Deloitte Tech Fast 50 company and a CIX Top 10 Growth company.

CEO Liran has also lined-up an impressive team of advisors and experts to advise BenchSci on life sciences research and development, organizational culture, and artificial intelligence and machine learning technology, including TCV Venture Partner Jessica Neal (former Chief Talent Officer at TCV portfolio company Netflix).

“BenchSci plays an important role in curating and contextualizing healthcare data to increase productivity in the preclinical research process,” says Jessica, “and I look forward to supporting Liran and his team as they continue to scale.”

Growth metrics and accolades aside, what also impressed us about BenchSci is Liran’s unwavering focus on fostering BenchSci’s culture. Liran believes the company’s distinguished culture is instrumental to its success, and he endeavors to build an inspiring, inclusive, and equitable work environment where employees are set up to thrive and have a meaningful career. Starting with Liran, it was clear during our diligence that BenchSci’s employees pursue continuous improvement and a high-degree of transparency and candor. The results speak for themselves – BenchSci has been named a certified Great Place to Work® and is a top-ranked organization on Glassdoor. We are excited to add Jessica Neal to BenchSci’s advisory board to help Liran continue to develop and grow the company’s culture as BenchSci scales through its next major inflection points.

“Our recent Series C raise enables us to build and deliver a next generation AI solution for global pharmaceutical companies that will enable scientists to exponentially improve their preclinical R&D work,” says BenchSci CEO Liran Belenzon. “We are a mission-driven organization intent on achieving success beyond success, and I’m excited that TCV recognizes our market-leading potential and has chosen to back our meteoric hypergrowth.”

We are off to the races in our partnership with Liran and the BenchSci team, and are incredibly excited to help build a category-defining, generational software company that drives productivity and efficiency in the preclinical research process, thereby bringing novel, life-saving medicines to patients faster.

***

The views and opinions expressed are those of the speakers and do not necessarily reflect those of TCMI, Inc. or its affiliates (“TCV”). TCV has not verified the accuracy of any statements by the speakers and disclaims any responsibility therefor. This blog post is not an offer to sell or the solicitation of an offer to purchase an interest in any private fund managed or sponsored by TCV or any of the securities of any company discussed. The TCV portfolio companies identified, if any, are not necessarily representative of all TCV investments, and no assumption should be made that the investments identified were or will be profitable. For a complete list of TCV investments, please visit www.tcv.com/all-companies/. For additional important disclaimers regarding this interview and blog post, please see “Informational Purposes Only” in the Terms of Use for TCV’s website, available at http://www.tcv.com/terms-of-use/