Anthropic-Backed Ode Buys Casper As AI Services Race Heats Up

Ode with Anthropic has acquired Casper Studios, a small AI services firm that helps companies deploy artificial intelligence into products, employee workflows and business processes. Financial terms were not disclosed. The deal adds another implementation-focused team to Ode, the standalone services company formed through a partnership involving Anthropic, Blackstone and Hellman & Friedman.

The acquisition looks small next to the enormous sums being spent on frontier AI models, data centers and chips. Yet the deal points to a part of the AI market that may prove just as consequential for enterprise buyers. Enterprise AI spending is shifting from experimentation toward deployment. Companies can now easily buy access to frontier models quickly, but turning those models into production systems is much harder. That requires integration work, governance, workflow redesign and engineering talent, creating a growing market for firms that can translate model capability into measurable business results.

The AI services market is shaping up to become a valuable layer of the generative AI market, and model companies, consulting firms, systems integrators and smaller engineering specialists are all moving toward the same opportunity. Owning the work that happens between a powerful model and the point where an enterprise can actually use it is becoming a new market battleground.

The Evolving AI Services Market

The next phase of enterprise AI is increasingly about implementation. Companies have spent the past several years testing large language models, building proofs of concept and giving employees access to tools such as Claude, ChatGPT, Microsoft Copilot and Gemini as well as an increasing array of open models. The harder work comes after those experiments. Enterprises need to connect AI to internal data, redesign workflows, establish governance, build custom applications and determine where automation produces measurable business value. That is the market Ode is trying to address.

Ode itself barely existed in its current form a few months ago. Anthropic announced the venture in May with Blackstone, Hellman & Friedman and Goldman Sachs, joined by investors including General Atlantic, Leonard Green, Apollo Global Management, GIC and Sequoia Capital. Anthropic said the company would place applied AI engineers beside Ode engineers to bring Claude into major operations at midsized businesses.

MORE FOR YOU

The venture formally became Ode in July. Its operational core came from Fractional AI, an applied AI services company acquired in May. Fractional founders Chris Taylor and Eddie Siegel became Ode’s CEO and CTO.

TechCrunch reported the business launched with roughly 100 engineers and $1.5 billion behind it. Ode operates on a “Claude-first” principle, giving Anthropic an unusually direct path from model development into corporate implementation. Its private equity backers bring something nearly as useful: companies to sell to. Their portfolio businesses can become prospective Ode clients.

Ode says it works closely with Anthropic's Applied AI organization and has direct access to technical expertise around Claude. Anthropic benefits when companies move from experimenting with Claude to using it in production. Services, in this case, can function as a distribution channel for the model.

Casper became an Anthropic Select Service Partner earlier this year. Anthropic has relationships with major consulting and technology services firms including Accenture, Deloitte, PwC, Cognizant and Infosys. It has committed significant resources to training partners that can implement Claude for corporate customers. Ode occupies another part of that market. Its pitch centers more heavily on applied engineers who work directly with clients to build production systems. With the acquisition of Casper Studios Ode with Anthropic aims to focus on employee workflows and operational processes.

Casper CEO and co-founder Jay Singh described the two companies as complementary.

“Ode goes deep on its clients’ top AI priorities, building production-grade applications for an organization’s hardest AI challenges,” Singh wrote in announcing the transaction. “Casper goes broad, helping deploy AI across teams and repeatable workflows.”

That distinction captures a growing problem in enterprise AI. Building an AI application is getting easier and easier, but getting an organization to use AI in dozens or hundreds of repeatable processes is much more difficult.

Enterprises have significant questions when it comes to making AI work in their organization. Where should AI be deployed first? Which processes should be redesigned? What data can the model access? How should permissions work? What happens when an agent makes a mistake? Which employees need to remain in the loop? How does a company calculate return on investment?

They are frequently organizational questions too. A company can easily access a frontier model , but that model cannot redesign a tax process, underwriting operation, customer support function or software engineering organization. That gap is creating a large market for AI services.

Forward-Deployed Engineering Is Becoming Part Of AI Sales

The emerging AI services model resembles the forward-deployed engineering approach that has become increasingly common among AI vendors. Enterprise software once relied heavily on salespeople and implementation partners, but AI is changing that formula.

Many AI products are too open-ended to sell like conventional software. A customer may know that Claude or another frontier model can perform sophisticated reasoning, coding or analysis. The customer may have no idea which use case deserves attention first.

The forward-deployed engineering role sits somewhere between software engineering, consulting, product development and technical sales. Engineers work directly with customers, sometimes for months, to turn a general-purpose AI model into something useful for a particular business. They help discover the use case, connect the model to company data, build the application and test it against real operational conditions.

OpenAI, Microsoft, Salesforce, Amazon Web Services and other vendors have been developing variations of this model. The Casper acquisition gives Ode people who have already been working on the broader organizational side of adoption, where AI moves beyond a flagship application and into ordinary employee workflows. That growing FDE army could become one of the larger defensible moats in enterprise AI.

AI labs need partners such as Accenture, McKinsey, Deloitte, PwC and Capgemini to reach large organizations. Yet they increasingly need their own deployment expertise too. This is resulting in a layered services market.

Large consulting firms can handle enormous transformation programs, change management and integration work. Smaller engineering-heavy firms have the depth of knowledge and speed to attack difficult AI applications, and specialists can focus on individual functions or industries. Model companies meanwhile can supply technical expertise close to the underlying technology.

In March, Anthropic committed $100 million to the Claude Partner Network, funding training, technical support and joint marketing for companies that deploy Claude. By June, Anthropic said more than 40,000 firms had applied to the program and more than 10,000 consultants had earned Claude certifications.

It has pursued another route through the traditional consulting industry. Anthropic and Accenture formed a dedicated business group in December, with plans to train about 30,000 Accenture professionals on Claude.

Anthropic is not alone in moving closer to services. OpenAI has been building its own deployment capabilities and deepening relationships with major consulting firms. Its efforts point toward the same conclusion. OpenAI launched the OpenAI Deployment Company in May and agreed to acquire applied AI consultancy Tomoro, bringing roughly 150 forward deployed engineers and deployment specialists into the company.

OpenAI committed more than $4 billion of initial investment to the venture. The investor and partner group includes TPG, Advent, Bain Capital and Brookfield, plus consulting firms Bain & Company, Capgemini and McKinsey.

In February, OpenAI had already expanded partnerships with BCG, McKinsey, Accenture and Capgemini around enterprise deployments. Reuters described the effort as a push to get companies beyond pilots and into core operations.

Supply and Demand Challenges for AI Consulting and Services

While many pundits and forecasters have called for a decline in consulting and advisory work due to the growing use of AI, this is not playing out in the current environment. AI is increasing demand for consulting at the same time that it reduces the labor required to perform consulting work.

IDC sees a large prize emerging. Its 2026 outlook estimates enterprise AI services could generate about $50 billion of additional IT consulting and systems integration spending over the following four years.

Traditional consulting and technology outsourcing businesses often make money by assigning teams of people to client engagements. Revenue has historically been connected, directly or indirectly, to labor. But now AI breaks that relationship.

A team using advanced coding models may finish work faster. An AI agent may automate research or documentation. A smaller group of senior engineers may accomplish work that once required a larger pyramid of junior employees. Clients are beginning to expect those productivity gains.

Reuters reported that major technology services customers in India are pressing providers for more output at lower prices. Persistent Systems CEO Sandeep Kalra said some customers want the same work for 25% to 30% less.

TCS CEO K. Krithivasan told Reuters that roughly 80% of contracts in parts of the company's business services operation are now tied to performance outcomes.

Reuters recently examined the problem in India’s $315 billion IT sector. Some firms are beginning to push toward what Reuters called “services-as-software,” mixing reusable AI platforms with service engagements. Smaller firms, such as AI and automation company qBotica have shifted towards subscription-based services arrangements as companies seek returns tied to productivity outcomes.

All of this is changing how services businesses might be valued. Investors may start paying less attention to employee count and more attention to revenue per employee, reusable technology, proprietary workflows, customer access and the ability to convert project work into repeatable products.

That shift is already visible in parts of the IT services industry. Smaller engineering firms have been winning work from larger incumbents in cases where customers value speed and technical depth more than sheer staffing capacity.

From this perspective, the Casper acquisition by Ode fits that pattern. Ode bought a team that already understands how companies are trying to turn Claude and related tools into operational workflows, rather than an increasingly commoditized labor pool.