| Constellation Research

SAP rolls out developer tools for Joule, ecosystem connections

Editor in Chief of Constellation Insights
Constellation Research

Larry Dignan is Editor in Chief of Constellation Insights at Constellation Research, where he leads editorial coverage focused on enterprise technology, digital transformation, and emerging trends shaping the future of business. He oversees research-driven news, analysis, interviews, and event coverage designed to help technology buyers and vendors navigate complex markets with clarity and context. ...

SAP advanced its plans for developer tools, SAP Build, the Joule roadmap and connecting to a broader ecosystem including a partnership with Snowflake.

At SAP TechEd 2025 in Berlin, the company outlined a series of AI-driven tools in SAP Build as well as a set of Joule Agents designed to enable developers to move faster.

Muhammad Alam, a member of the Executive Board at SP, said the innovations at SAP TechEd create a "unique flywheel of applications, data and AI put developers in the driver’s seat."

Specifically, SAP outlined the following:

Developers who use agentic AI platforms such as Cursor, Claude Code, Cline and Windsurf can now use SAP frameworks via SAP Build local Model Context Protocol (MCP) servers.
Visual Studio Code users will be able to use SAP Build via extensions.
The extension will be available later on the Open VSX Registry.
Joule Studio will get new tools to customize SAP ready-to-use agents as well as build new ones grounded in SAP business data.
SAP rolled out new Joule AI assistants to coordinate multiple agents across workflows, departments and applications for finance, supply chain and HR.
The company partnered with Snowflake on a new SAP Snowflake extension for SAP Business Data Cloud. The partnership gives joint customers the ability to move data bidirectionally.
SAP HANA Cloud knowledge graph ending can now automatically generate knowledge graphs.
SAP launched its first enterprise relational foundation model, which is focused on business outcomes. SAP-RPT-1, short for the first-generation Relational Pre-trained Transformer, can make predictions for common business scenarios including delivery delays and payment risk.

SAP Q3 mixed, outlook for 2025 at low end

Constellation Research analyst Holger Mueller said the SAP TechEd lineup was compelling. He said:

"SAP delivered a compelling keynote, showing how the SAP product teams work together for customer success on AI, application development and more - which was not always the case and good to see. With innovations on the Joule, SAP's agentic framework, advances on SAP Business Data Cloud, language driven and vibe code based development, ABAP models coming to on premises customers, SAP has shown significant innovation dynamic for its customer base. The question now is will it help SAP customers to tackle the task at hand - the upgrade to S/4HANA Cloud."

Here are the key points from Alam at SAP TechEd.

App–Data–AI Flywheel & ROI

SAP’s thesis: Real ROI comes from the seamless integration of applications, data, and AI, rather than isolated AI experiments.
SAP’s business suite spans finance, supply chain, HR, CRM, and more, giving it an advantage in creating a “global maximum” of value versus “local optimizations” seen in siloed systems.
Embedded AI in unified applications enables automation and insight across end-to-end business processes.

Assistants, Agents, and Human Productivity

SAP envisions AI as an assistive layer first—to make people smarter, faster, and more efficient—before evolving toward autonomous execution once trust is built.
Example: demand-planning and supply-planning agents collaborating autonomously across functions.
The “assistant” concept begins with human-in-the-loop confidence building, leading later to autonomous operations when maturity allows.

AI Hype vs. Real ROI

AI hype has inflated expectations. Many companies now realize value only comes from simplifying and structuring the equation—embedding AI in core workflows, not generating random apps.
95% of firms once reported no ROI (MIT study), but a newer study found 74% now see positive ROI—mainly those that adapt workflows and culture rather than just bolt on AI.
ROI perception varies: executives tend to report success, managers often don’t see it yet.

SAP’s Approach to AI Value Creation

SAP focuses on incremental AI adoption—enhancing existing roles and processes rather than replacing them.
Companies should start by making core roles (like AR clerks or billing agents) more efficient, then progress toward autonomous execution.
AI maturity follows a staged “value journey”: assistive → semi-autonomous → autonomous → deep research.

Developers, Product Managers & the AI Workforce

SAP employs ~40,000 developers but sees enough backlog for 200,000 developers’ worth of work—AI helps scale output, not replace people.
AI agents are multiplying productivity (7x–12x) in some teams by automating code generation, testing, and design.
SAP Build and Joule Studio Agent Builder will let developers and customers create AI agents in low-code/no-code environments.
Roles like product manager, QA, UX, designer will evolve but not disappear—AI will change team ratios and workflows.

Data Ecosystem & Partnerships

Business Data Cloud (BDC) launched in 2025 integrates with Snowflake and Databricks, allowing zero-copy data sharing.
SAP aims for open, governed, interoperable data management, letting customers combine SAP and non-SAP data seamlessly.

Organizational Design & Industry Collaboration

SAP is experimenting with “Dev Games of the Future” to rethink team structures and collaboration models for AI-augmented development.
Emphasis on evolving from people upward, not top-down reorganization.
SAP collaborates with large customers and institutions (e.g., Linux Foundation) on defining future role structures and standards.

ROI Beyond Headcount Reduction

SAP stresses AI ROI doesn’t equate to layoffs—it’s about growth, productivity, and redeployment of talent.
Customers view AI as a tool to expand capacity, not shrink the workforce.

Accountability & Measurement for AI

ROI and accountability come from embedding AI in structured business processes where performance can be measured (e.g., sourcing contracts closed, savings achieved).
SAP’s new Agent Topology Index maps and tracks agents (SAP and non-SAP) to measure outcomes and performance signals.

Quantum Computing Outlook

SAP is experimenting with quantum computing but sees it as post-2030 for commercial rollout.
Quantum will serve different use cases than AI, likely focusing on performance and optimization problems.

Data to Decisions Future of Work Innovation & Product-led Growth Tech Optimization Next-Generation Customer Experience Digital Safety, Privacy & Cybersecurity ML Machine Learning LLMs Agentic AI Generative AI Robotics AI Analytics Automation Quantum Computing Cloud Digital Transformation Disruptive Technology Enterprise IT Enterprise Acceleration Enterprise Software Next Gen Apps IoT Blockchain Leadership VR Chief Information Officer Chief Executive Officer Chief Technology Officer Chief AI Officer Chief Data Officer Chief Analytics Officer Chief Information Security Officer Chief Product Officer

Read more about SAP rolls out developer tools for Joule, ecosystem connections

Snowflake takes Snowflake Intelligence GA, launches developer tools, integrates with SAP BDC

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

Snowflake launched Snowflake Intelligence to general availability, outlined a set of new developer tools and forged a pact with SAP so Snowflake AI Data Cloud and SAP Business Data Cloud are interoperable.

The company said Snowflake Intelligence, outlined at Snowflake Summit earlier this year, is now generally available. Snowflake said that Snowflake Horizon Catalog and Snowflake Openflow, also announced at Snowflake Summit, are also generally available.

Snowflake Intelligence is designed to create agents that can operate via natural language, leverage structured and unstructured data and accelerate AI and machine learning pipelines. Customers embedded in Snowflake Intelligence include Cisco, Fanatics, Toyota Motor Europe and Wolfspeed.

According to Snowflake, prebuilt agents can be accessed through the Snowflake Intelligence interface or using Cortex Agents API. Cortex Agents orchestrate unstructured and structured data with LLMs including OpenAI GPT and Anthropic Claude.

Snowflake also outlined a suite of new development tools that feature a collaboration environment, open source integrations and data quality capabilities.

The developer tools include:

Cortex Code in private preview. Cortex Code is a revamped AI assistant in the Snowflake interface and helps customers understand usage, optimizations and fine tuning.
Enhancements to Snowflake Cortex AISQL, which is generally available. Developers can build AI pipelines in Snowflake Dynamic Tables.
Snowflake Workspaces for collaboration and direct Git Integration and VS Code Integration.

Under the SAP-Snowflake partnership, the companies said Snowflake's data and AI platform will be available as an extension for SAP Business Data Cloud. SAP launched Business Data Cloud via a partnership with Databricks.

SAP and Snowflake will enable the following:

Zero copy sharing between the two platforms.
Data and AI teams can work within a unified governance framework and harmonize SAP and non-SAP data.
Simplify AI governance, ground AI in enterprise knowledge bases and build tailored agents.
Leverage semantically rich data.
SAP Business Data Cloud can leverage Snowflake's AI, analytics, data engineering, marketplace and collaboration tools.

In addition to SAP Snowflake solution extension for SAP Business Data Cloud, the companies said the partnership includes SAP Business Data Cloud Connect for Snowflake, which enables bidirectional, zero copy data sharing. SAP Snowflake will be available in the first quarter with SAP Business Data Cloud Connect landing in the first half of 2026.

Data to Decisions snowflake Chief Information Officer

Read more about Snowflake takes Snowflake Intelligence GA, launches developer tools, integrates with SAP BDC

Snowflake launches Agent GPA, aims to grade your AI agents

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

Snowflake is looking to give your AI agents a GPA. While the company is grading the accuracy of AI agents, it's really evaluating goals, plans and actions (GPA) in an open source framework that reaches near human levels of error detection rates and localization accuracy.

The framework, called Agent GPA, was outlined at its Build conference. For enterprises deploying agentic AI, Snowflake's efforts are worth a look.

In a blog post, Snowflake's AI Research team said evaluating AI agents comes down to trust. Snowflake said:

"An agent’s answer may appear successful, but the path it took to get there may not be. Was the goal achieved efficiently? Did the plan make sense? Were the right tools used? Did the agent follow through? Without visibility into these steps, teams risk deploying agents that look reliable but create hidden costs in production. Inaccuracies can waste compute, inflate latency and lead to the wrong business decisions, all of which erode trust at scale."

Snowflake argued that current evaluation frameworks fall short because they focus on the final answer, not the process behind the answers. Here's a look at the Agent GPA framework. Agent GPA, outlined in a paper, is available in Truelens.

Snowflake's Agent GPA was the headliner among a set of items released by the company's research team.

Other items include:

Text-to-SQL V1.5, a specialized model that fuels Snowflake Intelligence, Snowflake's enterprise agent, by tackling the slowness, cost, and dialect issues of general LLMs. The specialized model makes text-to-SQL queries up to 3 times faster while maintaining accuracy.
New optimizations will be introduced for Cortex AISQL, a tool that integrates AI directly into SQL queries, enabling teams to analyze all data types and build flexible AI pipelines using familiar SQL syntax.
The Cortex AISQL enhancements improve AI operator efficiency and cost, featuring 2-8x more performant execution plans, 2-6x faster inference (at 90-95% accuracy), and a 15-70x reduction in execution costs and time through techniques like cost-aware optimization, adaptive model cascading, and query enhancements.

Data to Decisions Future of Work Innovation & Product-led Growth Tech Optimization Next-Generation Customer Experience Digital Safety, Privacy & Cybersecurity snowflake ML Machine Learning LLMs Agentic AI Generative AI Robotics AI Analytics Automation Quantum Computing Cloud Digital Transformation Disruptive Technology Enterprise IT Enterprise Acceleration Enterprise Software Next Gen Apps IoT Blockchain Leadership VR Chief Information Officer Chief Executive Officer Chief Technology Officer Chief AI Officer Chief Data Officer Chief Analytics Officer Chief Information Security Officer Chief Product Officer

Read more about Snowflake launches Agent GPA, aims to grade your AI agents

Celonis makes case for process, context in agentic AI, forges Databricks partnership

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

Celonis is looking to embed its process intelligence platform into agentic AI workloads and make it clear that AI agents sans process knowhow won't deliver enterprise value.

At Celosphere 2025 in Munich, Celonis outlined the following:

Celonis Process Intelligence additions that integrate data lakes without data duplication, tools to build more comprehensive digital twins of enterprises and blueprints for enterprise architecture.
A partnership with Databricks that integrates Delta Sharing to directly connect the Celonis Process Intelligence Platform with the Databricks Data Intelligence Platform.
And enterprise value use cases from the likes of Mercedes Benz, Uniper, Vinfast and 120 other "value champions" that have realized value of more than $10 million each.

Alex Rinke, co-CEO of Celonis, said the company is looking to help enterprises get value from their AI investments. "We give their AI the context it needs. We guide them to deploy it in the right places. And we enable them to make it work with everything else they’re doing," said Rinke.

Constellation Research analyst Mike Ni covered the Celosphere 2025 keynote live from Munich. He noted that Celonis was making the case that process intelligence should be the brain of enterprise AI. Ni added that Celonis touted an ecosystem that includes Wipro, Microsoft and now Databricks.

In many ways, Ni said Celonis is positioning itself as a "multi-dimensional Digital Twin of Operations" and the "architecture for contextual AI."

"While the hot term is not always understood think of it as a living business graph that feeds AI reasoners solving the very hard problem of context structuring," said Ni.

He added:

"The underlying message: stop treating AI as a science project. Make AI operational. AI suffers without context. Celonis wants to give it memory. Celosphere 2025 isn’t about process mining anymore, it’s about decision mining: finding, understanding executing where AI should act."

Here's a deeper dive on the Celonis news.

Celonis added to its Process Intelligence Platform and built out its operational digital twin features. The company said Celonis Data Core is generally available with the ability to integrate with Databricks and Microsoft data lakes.

The company said that the Process Intelligence Graph includes the ability to connect desktop actions to business processes with new task mining capabilities and AI-driven task discovery. Unstructured data types are also integrated.

Celonis also added enterprise architecture blueprints that give a live view of operations. The company also added more than 60 pre-built objects and events including assistants for data extraction and data modeling.

Other additions include:

New object-centric process mining (OCPM) capabilities to identify issues at process intersection points.
An orchestration engine that now coordinates AI agents for end-to-end processes.
Process Intelligence MCP Server.

As for the Databricks partnership, Celonis will leverage Databricks Delta Sharing for bi-directional integration. The integration means that joint customers won't need to move or copy data between platforms.

With the Databricks integration, Celonis can read live Databricks data and enrich it with business context and its Celonis Process Intelligence Graph to create digital twins. Process Intelligence from Celonis can then be fed back to Agent Bricks.

For Celonis to win over enterprise, it'll have to prove out its value in the data to decisions chain. Mercedes-Benz outlined how it uses Celonis for order-to-deliver, aftersales and quality management. Mercedes has integrated process intelligence with its AI and operations workflows.

Vinmar created a digital replica to automate order-to-cash, find distressed orders and remove friction from handoffs.

Data to Decisions Tech Optimization Innovation & Product-led Growth Future of Work Next-Generation Customer Experience Revenue & Growth Effectiveness Chief Information Officer

Read more about Celonis makes case for process, context in agentic AI, forges Databricks partnership

Clorox ERP implementation hangover dings fiscal Q1

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

Clorox still has an ERP implementation hangover as the company reported a 19% decline in sales "primarily driven by lower shipments related to the ERP transition."

The company reported fiscal first quarter earnings of 65 cents a share on revenue of $1.43 billion. Adjusted earnings were 85 cents a share, down by 54% from a year ago.

Clorox had been on an SAP ERP system that was more than two decades old. It set out to upgrade SAP in 2021 and the final tab will range from $560 million to $580 million. As previously reported, Clorox wrapped up the ERP upgrade and was poised for better margins ahead.

However, Clorox executives said the company shipped more inventory for a buffer as it transitioned to the new ERP system. When retailers ran out of stock, Clorox ran into procurement issues.

Time to Introduce Hi-Fi for Enterprise Processes

Here's what happened via CEO Linda Rendle and CFO Luc Bellet in prepared remarks.

"At the end of last fiscal year, we shipped roughly two weeks of inventory ahead of consumption as retailers built stock in preparation for this transition. As those inventories were drawn down this quarter, we expected net sales to decline by approximately 17% to 21%. While the ERP systems cutover proceeded smoothly, we encountered some challenges in order fulfillment that led to temporary out-of-stocks."
"Based on early estimates, we projected and communicated in early September that our first-quarter net sales were likely to come in at the low end of our guidance range. Ultimately, this quarter’s results exceeded these expectations for two key reasons: first, the impact of fulfillment disruptions was less significant than anticipated due to stronger-than-expected recovery in September; and second, we shipped ahead of consumption for some second-quarter merchandising events."

Clorox said it didn't help that the economy and consumers are under pressure. Clorox said consumers are showing "value-seeking behaviors across all income segments."

Now Clorox is focusing on regaining share from the temporary out of stocks. Executives also noted that the issues have been addressed and fulfillment processes have stabilized. Clorox concluded:

"Our transformation and ERP implementation strengthens our digital backbone and positions us to unlock meaningful operational efficiencies, margin expansion and superior value for our consumers. In an environment where consumer dynamics are changing rapidly, our new tools are allowing us to reach them in new and innovative ways, improving the returns on each dollar invested. With each wave of implementation, we’re gaining sharper insights and deeper operational visibility — enabling faster, smarter, and better execution."

Data to Decisions Tech Optimization Digital Safety, Privacy & Cybersecurity Innovation & Product-led Growth Future of Work SaaS PaaS IaaS Cloud Digital Transformation Disruptive Technology Enterprise IT Enterprise Acceleration Enterprise Software Next Gen Apps IoT Blockchain CRM ERP CCaaS UCaaS Collaboration Enterprise Service Chief Information Officer Chief Technology Officer Chief Information Security Officer Chief Data Officer

Read more about Clorox ERP implementation hangover dings fiscal Q1

Palantir US commercial revenue jumps 121% in Q3, ups outlook

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

Palantir continued to land commercial accounts as its third quarter results handily topped expectations. The company’s US commercial revenue was up 121% from a year ago.

The company reported third quarter earnings of $476 million, or 18 cents a share. Non-GAAP earnings were 21 cents a share on revenue of $1.18 billion, up 63% from a year ago.

Wall Street was expecting Palantir to report earnings of 11 cents a share (17 cents a share non-GAAP) on revenue of $1.09 billion.

During the quarter, Palantir announced partnerships with Hadean, Lear, Lumen Technologies and SOMPO. Palantir has increasingly been competing with more traditional enterprise software vendors and winning accounts via its data ontology and ability to deploy AI.

Based on the Rule of 40, Palantir blew that away with a score of 114%. CEO Alex Karp said the company’s US business growth was 77%. Most of Palantir’s revenue is focused on the US as well as western allies.

By the numbers:

US revenue was $883 million in the third quarter.
US commercial revenue was $397 million with US government revenue of $486 million, up 52% from a year ago.
The company said it closed $2.76 billion of total contract value in the third quarter, up 151% from a year ago.
Customer count was up 45% from a year ago.
Palantir had $6.4 billion in cash and equivalents.
In the third quarter, Palantir closed 204 deals worth at least $1 million, 91 worth at least $5 million and 53 worth at least $10 million.

As for the outlook, Palantir raised its outlook and projected 2025 revenue between $4.396 billion and $4.4 billion. Commercial revenue will top $1.43 billion. Fourth quarter revenue will be between $1.327 billion and $1.331 billion.

In a shareholder letter, CEO Alex Karp delivered his usual complement of strong quotes. He said:

“This remains the beginning, the first moment of a first chapter.”
“It is worth remembering that the business is now producing more profit in a single quarter than it did in revenue not long ago. This ascent has confounded most financial analysts and the chattering class, whose frames of reference did not quite anticipate a company of this size and scale growing at such a ferocious and unrelenting rate. Some of our detractors have been left in a kind of deranged and self-destructive befuddlement.”
“Our partners in the United States—the earliest and most voracious adopters of the novel language models that are presently reordering human life and of the Ontology that allows them to effectively operate—understand how significantly the terrain beneath us all has shifted.”

Data to Decisions Innovation & Product-led Growth Chief Executive Officer Chief Financial Officer Chief Information Officer

Read more about Palantir US commercial revenue jumps 121% in Q3, ups outlook

AI in the Boardroom: Redefining Consulting, Talent & Trust

Liz Miller

Vice President & Principal Analyst
Constellation Research

About Liz Miller: Liz Miller is Vice President and Principal Analyst at Constellation, focused on the org-wide team sport known as customer experience. While covering CX as an enterprise strategy, Miller spends time zeroing in on the functional demands of Marketing and Service and the evolving role of the Chief Marketing Officer, the rise of the Chief Experience Officer, the evolution of customer engagement, and the rising requirement for a new security posture that accounts for the threat to brand trust in this age of AI. With over 30 years of marketing experience, Miller offers strategic guidance on the leadership, business transformation, and technology requirements to deliver on today’s CX strategies. She has worked with global marketing organizations to transform everything from…...

"All business is personal" — Jay Laabs, CEO & Founder of Spaulding Ridge, shares how deep client relationships, trust, and strategic upskilling are driving AI-powered transformation in consulting. From boardroom pressures to leadership imperatives like ‘freedom within a framework,’ discover how AI is changing not just business models, but company culture itself. This is for anyone navigating the evolving landscape of AI, data, and innovation. Watch this live interview with Liz Miller during Dreamforce 2025!

Future of Work Innovation & Product-led Growth Next-Generation Customer Experience Tech Optimization

Read more about AI in the Boardroom: Redefining Consulting, Talent & Trust

On ConstellationTV <iframe width="560" height="315" src="https://www.youtube.com/embed/9yNEUaB9g6E?si=VosCa9ijM6TSrXHi" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>

MongoDB names CJ Desai CEO

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

MongoDB said it has appointed CJ Desai, an alum of Cloudflare and ServiceNow, as CEO effective Nov. 10.

In a statement, the company said current CEO Dev Ittycheria will retire from a full-time role but will stay on MongoDB's board.

It was evident that Desai was going to land a CEO gig when Cloudflare reported its earnings. Cloudflare said Desai was resigning as president of product and engineering to take a CEO role at a visible publicly traded company.

Desai has held technology leadership roles at Cloudflare, ServiceNow and held roles at EMC and Symantec over the last 25 years. Desai's highest profile role was at ServiceNow where the company's annual revenue went from $1.5 billion to $10 billion. Desai resigned from ServiceNow in 2024 after violating company policy.

In a statement, Desai said:

"MongoDB has long been the partner of choice for building applications that transform businesses, and now it is uniquely positioned to power the next wave of AI-driven applications. My directive is clear: by staying relentlessly close to customers, delivering category-defining products, and executing at scale, we can seize the enormous opportunities ahead."

Desai takes over for Ittycheria as MongoDB is performing well. MongoDB reported a strong second quarter and raised its outlook.

MongoDB said its third quarter results will be at the high end of its earnings and revenue guidance. MongoDB said it will report earnings on Dec. 1.

Scale and strategy

Wall Street analysts asked Desai about strategy, go-to-market scale and the architecture behind AI workloads.

Desai said MongoDB can further scale. A few key takeaways include:

"MongoDB has long been the partner of choice for applications that transform businesses, and I believe the company is exceptionally positioned to power the next wave of AI applications."
Desai said his experience in scaling companies will be useful to MongoDB. Desai said MongoDB’s market motion will be to expand from developers to the C-suite.
"MongoDB participates in a very large growing market, and it has also a very clear architectural advantage to become part of the architecture for modern workloads. There are so many workloads being changed to leverage AI. We have reached this inflection moment where Mongo can truly become the heart of this next phase of architecture. MongoDB architecture was not force fitted for AI workloads. It existed for AI workloads."
Desai said that AI workloads are rewiring database decisions. “I think the rearchitecture is happening right now, and it is our job as MongoDB to ensure that customers understand the why and how we are ideally suited for those applications,” he said.

Desai, who started his career at Oracle, was asked about MongoDB's positioning given he oversaw the ServiceNow acquisition of RaptorDB. He was also asked about the limitations of Postgres. He said:

"When I truly look at the technology of MongoDB, and I spend some time truly evaluating the recent innovations from MongoDB on 8.0 and post 8.0 releases, what I would say is that this is the right database to build or modernize an application for. Relational databases tend to be very rigid and do not handle the unstructured data really well. When you think about AI applications of the future, it will still be both structured and unstructured data. And as the business changes, MongoDB would still be on the right side of that equation. Many of our customers want to scale that application for AI workloads."

Regarding Postgres, Desai said ServiceNow wanted to maintain the SQL functionality and structured data approach. "Our job with the team is to continue to make sure that MongoDB is top of mind, understand the advantages of MongoDB and to be flexible for scale out," he said.

Desai was also asked about competition from cloud providers. He said MongoDB's launch of Atlas nailed the cloud transition and ability to be agnostic. He said that MongoDB's multicloud approach is a moat. "Multicloud is here to stay," said Desai, who noted recent cloud outages. "MongoDB with a cloud agnostic architecture and how seamlessly it can work in a multi cloud environment is a competitive moat for MongoDB."

He added that MongoDB has tailwinds in database migrations to the cloud as well as AI workloads. Desai said international, notably India, also remains a growth area for the company. Desai will talk more about MongoDB strategy on the December 1 earnings call.

Constellation Research analyst Holger Mueller said:

"This brings Desai's career to a full circle -- 30 years ago he and his team built databases at Oracle and now he is at the helm at one of the few Oracle competitors left standing. This is good news for MongoDB customers and the overall database market that has atrophied. Data remains critical in the AI era and the most coveted data is in documents."

Data to Decisions mongodb Chief Information Officer Chief Data Officer

Read more about MongoDB names CJ Desai CEO

OpenAI inks $38 billion deal with AWS, starts renting GPUs immediately

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

OpenAI said it has signed a $38 billion agreement to use AWS and its Amazon EC2 UltraServers immediately. Under the deal, OpenAI will access hundreds of thousands of NVIDIA GPUs and likely CPUs for AI agents.

For AWS, the OpenAI deal is another big win. AWS reported 20% revenue growth in the third quarter and launched its Project Rainer cluster for Anthropic. AWS will announce Trainium3 at re:Invent 2025.

The big takeaways from this OpenAI-AWS deal may be the following:

OpenAI and Anthropic are diversifying compute with the latter more likely to leverage custom silicon from hyperscalers. In the AWS case, Anthropic is using Trainium2. For Google Cloud, the Anthropic deal is about the TPUs. OpenAI is diversified across Microsoft Azure, Google Cloud, Oracle Cloud Infrastructure and now AWS. See: OpenAI, Anthropic increasingly diverge as strategies evolve
OpenAI is more of a Nvidia shop, but it'll be interesting to see whether it winds up diversifying into AWS custom silicon--especially for CPUs.
The AWS-OpenAI deal is another data point in the AI industry's circular economy. OpenAI has more than $1 trillion in future commitments. However, OpenAI appears to be paying AWS immediately.

In the statement, AWS and OpenAI said:

"OpenAI will immediately start utilizing AWS compute as part of this partnership, with all capacity targeted to be deployed before the end of 2026, and the ability to expand further into 2027 and beyond."

Other key points:

The deal is focused on Nvidia GPUs but can expand to "tens of millions of CPUs" for agentic workloads. Inference will be a huge market as AI agents scale.
The OpenAI infrastructure on AWS will include clusters of Nvidia's GB200s and GB300s via Amazon EC2 UltraServers.
AWS said the OpenAI deal includes inference for ChatGPT as well as training new models.
OpenAI and AWS started to work together when the AI company put its open weight foundation models on Amazon Bedrock.

Data to Decisions Tech Optimization Innovation & Product-led Growth Future of Work Next-Generation Customer Experience Digital Safety, Privacy & Cybersecurity openai amazon ML Machine Learning LLMs Agentic AI Generative AI Robotics AI Analytics Automation Quantum Computing Cloud Digital Transformation Disruptive Technology Enterprise IT Enterprise Acceleration Enterprise Software Next Gen Apps IoT Blockchain Leadership VR Chief Information Officer Chief Executive Officer Chief Technology Officer Chief AI Officer Chief Data Officer Chief Analytics Officer Chief Information Security Officer Chief Product Officer

Read more about OpenAI inks $38 billion deal with AWS, starts renting GPUs immediately

TD SYNNEX CIO Kristie Grinnell on adopting AI, tech debt and change management

Larry Dignan

Editor in Chief of Constellation Insights
Constellation Research

TD SYNNEX CIO Kristie Grinnell outlined the tech distributor’s AI plans and how change management is key to driving revenue growth and productivity.

Here’s a look at the takeaways from my conversation with Grinnell, a Constellation Research BT150 member.

TD SYNNEX’s vision is to become an “AI-fluid company.” Grinnell described how the distributor’s AI strategy is built around empowering every employee to use AI tools intelligently rather than isolating them to specialists:

“We want to grow our business with the use of AI so it will augment our workforce. We envision a hybrid workforce, but we also envision that the future of distribution will become more digital as well. Now we don't think that the human will ever completely be out of the loop, but it’s imperative that we become what we call an AI fluid company where everybody can speak AI, know when to use an agent, know when to use Gen AI, know when to use machine learning and know when to even use a bot versus that toolkit.”

AI is central to both revenue growth and operational efficiency. Grinnell emphasized that AI doesn’t just save money—it’s equally a revenue driver:

“We feel it has to work for us to grow and our customers want the same experience. They want TD SYNNEX employees to be able to walk them through where they are on that journey digitally. So, there's a revenue piece to that. But we absolutely see that we're going to have to find the efficiencies in our own company so that as we grow that revenue, we're growing our profit as well.”

The “AI factory” concept will tackle tech debt and enable agility. Grinnell framed TD SYNNEX’s modernization around an “AI factory” that turns agents into the new layer of applications while gradually eliminating legacy complexity:

“We want to create what everybody's calling the AI factory. That factory is going to have to be able to talk to the data as well as the applications as inputs and create essentially new agents that become the new applications, overriding some of that tech debt that we have. Our goal would be that one by one, those agents will help us start to take a little bit of the tech debt away.”

Data discipline begins with focused domains, not “boiling the ocean.” She explained that effective AI depends on pragmatic data strategies—starting small, cleaning up legacy complexity, and proving value domain by domain:

“The approach we're using is that you just have to pick a domain as if this is where we're going to start and then we can go forward. We have so much data. It can be messy. We just have to really focus on one area first.”

The greatest challenge is change management and workforce education. Grinnell said success with AI depends more on people and learning than on technology alone:

“Every employee has an opportunity now that they didn't before. It's really not about the technology at this point. It's about the education, the knowledge and the know-how of your employee base. For us, this is more the biggest change management exercise we've ever led, rather than the biggest IT project we've ever led.”

More from CCE 2025:

Data to Decisions Future of Work New C-Suite Next-Generation Customer Experience BT150 Chief Information Officer

Read more about TD SYNNEX CIO Kristie Grinnell on adopting AI, tech debt and change management

Subscribe to

SAP rolls out developer tools for Joule, ecosystem connections