Nvidia backs Indian startup helping companies run AI models with fewer GPUs
By CelebUD Newsroom —
In an era where the generative artificial intelligence boom is placing unprecedented pressure on global computing infrastructure, semiconductor heavyweight Nvidia has invested in Simplismart, an Indian startup developing software designed to optimize artificial intelligence workloads and reduce the number of expensive graphics processors companies need to run them.
The Bengaluru-based enterprise secured $14 million in an all-primary Series B financing round led by Dallas Venture Capital. Alongside Nvidia, the round attracted participation from prominent venture firms Accel and Shastra VC, as well as consumer electronics manufacturer Micromax. The capital infusion establishes Simplismart's post-money valuation at approximately $100 million, marking a major milestone for the four-year-old infrastructure outfit.
Tackling the Astronomical Costs of AI Workloads
As enterprise adoption of generative models accelerates worldwide, businesses have directed billions of dollars toward high-performance graphics processing units (GPUs) to power real-time inference—the process of running live queries and generating content through trained machine learning models. However, operating these systems reliably, consistently, and economically at massive commercial scale has proven to be a persistent technical hurdle.
Different enterprise workloads require distinct trade-offs across response latency, compute expense, and operational throughput. Configuring these systems often demands intricate, continuous technical tuning. Simplismart resolves this operational bottleneck by automating the deployment pipeline, positioning its proprietary software layer directly between artificial intelligence models and the underlying hardware infrastructure.
The company's software automates how systems are provisioned and operated, adjusting execution based on real-time traffic without forcing software engineers to manually calibrate underlying compute instances. According to the firm's leadership, the prevailing industry obsession with raw speed or basement pricing overlooks practical enterprise realities.
“Nobody actually needs the fastest or the cheapest inference,”
stated Amritanshu Jain, co-founder and chief executive officer of Simplismart. He explained that commercial clients instead demand solutions that balance operational velocity and financial expenditure in alignment with their targeted use cases.
“Where we tend to win is where people have to be more stringent on their SLAs and TCO,”
Jain added, highlighting that customers frequently select the platform when navigating rigid service-level agreements and stringent total cost of ownership mandates.
Technological Capabilities and Verified Cost Reductions
Simplismart provides flexible deployment frameworks: organizations can utilize the company's native managed platform or run the optimization engine directly within their own private clouds or on-premises data centers. The platform incorporates comprehensive operational tooling, including capabilities for model fine-tuning, dynamic auto-scaling to absorb shifting user traffic, and continuous performance observability.
Additionally, the architecture enables third-party cloud service providers to package and host their own serverless inference products atop existing hardware stacks. Indian cloud infrastructure firm Yotta, for instance, has incorporated Simplismart’s engine to run the serverless AI inference tier inside its Shakti Studio suite.
Enterprise clients in the media generation space have reported measurable operational efficiencies after adopting the technology:
- InVideo: The video-generation platform reduced its GPU serving expenditures by 56% following integration.
- HeyGen and Higgsfield: Advanced synthetic video developers utilizing the software to handle high-demand media processing.
- Ema: An enterprise artificial intelligence firm operating complex operational pipelines via the platform.
The Paradox of Chipmakers Funding Hardware Efficiency
Nvidia’s direct financial backing of an enterprise whose core value proposition involves running workloads on fewer GPUs might initially appear counterproductive to a chip manufacturer's commercial interests. However, company leadership views the dynamic as mutually beneficial rather than adversarial.
Jain pointed out that lowering operational barriers and driving software efficiency makes enterprise hardware investments far more attractive, ultimately stimulating wider commercial adoption of artificial intelligence applications. Beyond capital support, Nvidia has initiated deep technical collaborations with Simplismart. Software engineers from the startup are actively interfacing with Nvidia’s solutions architects, focusing heavily on technical integrations surrounding Nvidia's Dynamo inference software. Nvidia has additionally facilitated direct commercial introductions, connecting the Bengaluru startup with prospective enterprise clients and ecosystem partners.
Rapid Financial Expansion and Market Dynamics
Simplismart has experienced rapid commercial growth over the past two quarters. The startup's annualized revenue run rate surged 14-fold over the preceding six months, reaching an eight-figure sum based on active customer workload consumption rather than fixed recurring subscriptions, excluding capital passed through directly to infrastructure suppliers.
| Metric | Company Milestone / Detail |
|---|---|
| Total Series B Capital Raised | $14 Million |
| Company Valuation | Approximately $100 Million |
| Lead Investor | Dallas Venture Capital |
| Key Participating Investors | Nvidia, Accel, Shastra VC, Micromax |
| Six-Month Revenue Growth | 14x increase to eight-figure run rate |
| Current Workforce | 58 employees |
| Primary Market by Revenue | United States (over 50%) |
While the company maintains a larger volume of individual client accounts within India, more than half of its aggregate revenue originates from the United States. To capitalize on this trajectory, Simplismart intends to use the Series B proceeds to accelerate hiring, establish a deeper operational footprint in the American market, and fund continuous research and development for its software platform.
Simplismart was founded in 2022 by Jain, a former software engineer at Oracle, and Devansh Ghatak, who previously engineered systems at Google. The co-founders first connected as undergraduate students at the Birla Institute of Technology and Science (BITS), Pilani, eventually establishing the startup based on their practical experience engineering and scaling large machine learning systems. Today, the organization employs 58 staff members across its offices.
The formal round closure also resolves recent market rumors. Following an earlier report from Indian financial daily The Economic Times alleging that the company was seeking between $40 million and $50 million at a valuation ranging from $300 million to $400 million, Jain confirmed that the current Series B round is entirely finalized, will not be extended, and that the company is not currently soliciting further investment.
What this shift means for emerging market innovation
Much of the global artificial intelligence conversation has centered on the United States and massive capital-intensive data centers. Yet the success of an Indian engineering team in attracting backing from global giants like Nvidia and Accel offers a compelling blueprint for other emerging tech hubs, including Nigeria, Kenya, and South Africa.
Emerging markets excel at engineering under constraints. When local environments grapple with bandwidth, expensive imported hardware, and strict spending limits, developers naturally prioritize resource efficiency. As the AI industry enters an era where making models lighter and cheaper is valued at hundreds of millions of dollars, African engineers building lean infrastructure solutions have a genuine global market to target.
Takeaway: Rather than simply importing Western AI models, can local developers and university incubators focus on building the optimization, fine-tuning, and infrastructure tools that help resource-constrained markets deploy modern technology sustainably?
Conclusion
Simplismart’s expansion highlights a decisive shift in the artificial intelligence sector: as the baseline capacity of foundational models matures, commercial viability hinges on driving down the cost of real-time execution. While hyperscale cloud providers such as Amazon Web Services, Google Cloud, and Microsoft Azure offer standard hosting tools alongside internal software released by Nvidia itself, specialized optimization startups are carving out valuable territory by addressing granular enterprise constraints. By securing the financial and technical backing of the world's most dominant AI hardware supplier, Simplismart is well-positioned to expand its footprint across North American enterprises seeking sustainable compute economics.
Reporting based on coverage by TechCrunch.
Category: Technology