DoiT Buys AI FinOps Startup Attribute, Launches AI Token Cost Management Product

‘We are now the only vendor who can provide real-time, kernel-level cloud shared cost attribution that doesn’t depend on anyone reporting anything,’ DoiT’s Field CTO Amit Kinha tells CRN.

Businessmen making handshake with partner, greeting, dealing, merger and acquisition, business cooperation concept, for business, finance and investment background, teamwork and successful business

DoiT has acquired startup Attribute to make AI, GPU and agentic AI costs more transparent than ever before—marking the company’s fifth acquisition over the past 18 months.

The Santa Clara, Calif.-based cloud channel standout is hitting the M&A ground running by launching a new product based on Attribute innovation that manages AI spending by correctly attributing AI spend across tokens, model requests, and GPU usage with zero instrumentation.

“We are now the only vendor who can provide real-time, kernel-level cloud shared cost attribution that doesn’t depend on anyone reporting anything,” Amit Kinha, DoiT’s field chief technology officer, tells CRN. “We are the only one that can observe AI cost from the source that triggered it, be it a person or agent.”

DoiT’s CTO said AI spending can quickly get out of hand. Thankfully for its AI and cloud customers, DoiT is launching a new product called ‘Attribute,’ based off its recently acquired Attribute technology.

“AI spend is more volatile, more decentralized, less responsive to good governance, and costs can explode in real time—especially when agents and sub-agents get involved and take independent action with shared GPUs, shared model accounts, and LLM gateways that older tools were never designed to see into,” said Kinha.

Kinha said Attribute tells clients the difference between human-triggered spend and agent/sub-agent spend; attributes those workflows and costs to the proper individual and project that triggered them; and gives both finance and practitioners teams that information in real time.

“We weren’t content to just help them better understand their AI bills after the fact, when the damage was done,” DoiT’s CTO said.

Financial details of the acquisition were not disclosed.

‘Every Token, Model Call And GPU Cycle Traces Back’

DoiT was seeking to acquire a company that could truly drive AI return-on-investments (RIO) for businesses.

The company’s research found that nearly 80 percent of enterprises had AI cost overruns in the past year, and the organizations with the most sophisticated governance had the highest overrun rates.

“We decided this problem requires a distinct approach and technology architecture built for the realities of AI, not just an iterative extension of our cloud cost platform,” said Kinha.

DoiT identified that startup Attribute, which is based in San Francisco and Israel, had built exactly the capability it needed to achieve their goals. DoiT moved quickly to acquire Attribute, which was founded in 2023, and build the technology into DoiT’s portfolio.

“Now, rather than aggregating what the AI providers report after the fact, our platform uses a lightweight kernel-level sensor to observe exactly what runs, in real time, at the kernel level, with no instrumentation, code changes, SDKs, or tagging,” Kinha said.

A lightweight eBPF sensor installs in about 15 minutes and starts producing per-customer, per-feature, and per-agent token economics the same day.

The sensor observes real consumption inside the operating system kernel and maps every unit of GPU, CPU, API call, memory, network, and I/O back to the process, container, pod, and request responsible.

It also identifies each outbound call to a managed model API and joins it with provider cost data, so token spend on Anthropic, OpenAI, Google Gemini, and AWS Bedrock lands on the workload, tenant, and agent that drove it rather than on a shared account no one can break down.

“Every token, model call and GPU cycle traces back to the workload, the feature, the agent, and the individual that triggered it,” he said. “For the first time, real-time AI spend attribution gets the right information to the right people fast enough to do something about it before the next bill arrives.”

DoiT’s Five Recent Acquisitions

DoiT has acquired five companies over the past 18 months.

In January, DoiT acquired Select, a data platform optimization company focused on Snowflake cost and performance.

In October 2025, DoiT bought multi-cloud security posture and compliance platform provider CloudWize.

Another recent acquisition included DoiT’s purchase of LiveDiagrams, a cloud application and architecture mapping solution provider.

In early 2025, DoiT acquired PerfectScale, an automated Kubernetes optimization and governance platform provider.

‘This Is A Deliberate, Strategic Shift’ For DoiT

Kinha said the acquisition of Attribute and launch of DoiT’s new Attribute product is “a deliberate, strategic shift” for his company.

“We’re applying the same approach that made us a leader in cloud cost management to AI tokenomics, and establishing ourselves as an innovative player in what we call automated AI cost attribution, where we can tie every token of spend to the work and the outcome it produced,” he said.

With Attribute, DoiT clients can attribute every single token to the customer, feature, product or team that drove it, with zero instrumentation.

“This makes us a more complete partner to the large enterprise and upper mid-market companies we already serve, and it gives prospects a reason to switch that the rest of the market can’t match,” said Kinha.