Ema lands $77M to scale AI agent platform competing with enterprise software

Ema, which builds teams of AI agents designed to streamline corporate workflows in human resources, information technology, and finance departments, has closed a $77 million Series B funding round. The investment, led by Bengaluru-based Creaegis with participation from returning backers Accel, Section 32, and Prosus, brings the company's total capital raised to $140 million and represents more than a fourfold increase in valuation from its previous 2024 round. (The startup has chosen not to disclose its current valuation.) The round consisted solely of primary equity with no debt or secondary components, according to the company.

The funding reflects a broader shift in enterprise technology spending, where artificial intelligence is beginning to capture market share traditionally reserved for conventional software platforms and services providers. Venture-backed startups, frontier AI research labs, and established software vendors are all competing to claim a portion of this spending.

Ema was established in 2023 by Surojit Chatterjee, a former executive at Google and Coinbase, alongside Souvik Sen, who previously worked at Okta. The company's core offering centers on what it terms "AI employees"—systems that orchestrate multiple AI agents to execute complex, multi-step business processes across an organization's existing software stack, rather than addressing isolated tasks individually.

Chatterjee envisions this architecture eventually diminishing enterprise reliance on conventional software products, particularly those delivered as software-as-a-service offerings. The startup initially integrates with a company's current applications, after which customers may progressively reduce or eliminate their dependency on certain products altogether.

Many of our customers are already on the way to replace [large SaaS applications] completely, removing dependency on them, because they are mostly becoming like a database

Surojit Chatterjee

Frontier AI labs as accelerators, not rivals

Leading artificial intelligence companies have recently intensified their push into the enterprise sector where Ema operates. Anthropic has broadened its efforts to integrate Claude into companies' essential operations, spanning financial and legal processes. OpenAI has similarly assembled teams of field-deployed engineers embedded with customers to operationalize AI systems.

Chatterjee does not characterize frontier AI research organizations as direct competitive threats. He explained that Ema's platform can leverage more than 150 different models—encompassing both cutting-edge and open-source options—while the company concentrates on the specialized knowledge, system connections, and process coordination required to fully automate end-to-end business operations.

Progress in frontier models is actually very beneficial to us

Surojit Chatterjee

Ema's strategy is demonstrating commercial viability. The company operates more than 50 active enterprise agreements and serves over 1 million active enterprise users, having processed more than 5 million actions and queries. Its customer roster includes NTT DATA, Hitachi, ADP, PwC, Google, KPMG, Wipro, and Microsoft.

Revenue has expanded 50-fold over the previous two years, with revenue bookings surpassing $150 million. Chatterjee clarified that the bookings figure encompasses the aggregate value of multiyear contracts, including two- and three-year agreements, rather than representing annualized recurring revenue. He declined to share the company's current annualized revenue run rate.

According to Chatterjee, more than 90% of Ema's customer base has broadened their usage beyond their initial deployment, with certain organizations implementing the technology across dozens of separate workflows. The company maintains a net dollar retention rate of approximately 180%, indicating that existing customers substantially increase their spending with Ema as time progresses.

Ema's ambitions extend beyond the software product itself. Chatterjee noted that artificial intelligence can assume portions of the implementation, integration, and advisory services that organizations have conventionally outsourced to IT services firms for enterprise software deployments.

A lot of the services companies are working with us. They are also dramatically changing or disrupting their own business models because they understand the human-forward model may not be the best model going forward.

Surojit Chatterjee

Despite assuming responsibilities traditionally performed by software and services organizations, Ema has sustained gross margins near 80%. Chatterjee noted that the company requires diminished human support as its AI systems improve through real-world deployments, which supports margin expansion over time.

Ema does not bill customers according to software user seats or the volume of AI tokens consumed. Rather, Chatterjee explained, pricing correlates to task completion and business results achieved.

The company intends to deploy much of the new capital toward strengthening its commercial operations, particularly in sales and marketing, having concentrated its early years on product development. The Mountain View-based organization has grown to nearly 200 staff members and maintains offices in Bengaluru, London, and Vancouver.

Ema has concentrated its efforts on serving customers in North America and Western Europe to date. The company now intends to enter additional geographic markets throughout the coming year, with particular focus on Asia-Pacific, South America, and select Middle Eastern regions.