Nvidia Backs Indian AI Infrastructure Startup Simplismart in $14 Million Funding Round to Optimize Model Deployment

As the global artificial intelligence boom places unprecedented strain on computing capacity and hardware availability, chip giant Nvidia has thrown its weight behind Bengaluru-based startup Simplismart. The company has secured $14 million in a Series B funding round, betting heavily on software designed to help corporations run AI models more efficiently, sustainably, and, in many cases, with a reduced reliance on expensive graphical processing units.

The all-primary funding round was led by Dallas Venture Capital, with participation from prominent global and regional backers including Accel, Shastra VC, Micromax, and Nvidia itself. According to the startup, this new infusion of capital values the four-year-old enterprise at approximately $100 million.

The investment arrives at a critical juncture for the technology sector. Enterprises worldwide are dedicating massive portions of their budgets to acquiring and operating high-performance GPUs to power modern AI applications. However, deploying these complex models reliably and economically at scale remains a persistent logistical and financial hurdle. Different workloads and business applications demand distinct, delicate trade-offs between processing speed, operational cost, and overall performance. Deploying and maintaining AI infrastructure to meet strict corporate requirements typically demands intricate configuration and ongoing adjustments.

Simplismart positions its proprietary software as a bridge that automates much of this operational heavy lifting. Instead of forcing corporate customers to manually manage complex low-level configurations, the startup’s platform tailors how machine learning models run on existing computing infrastructure.

"Nobody actually needs the fastest or the cheapest inference," said Amritanshu Jain, co-founder and chief executive officer of Simplismart, in an interview discussing the operational realities facing modern enterprises. Jain explained that companies instead require an infrastructure balance—running AI models at speeds and costs that make practical sense for their specific, real-world applications.

Operating as an intelligent layer that sits securely between AI models and the underlying computing infrastructure, Simplismart’s software dynamically adjusts how models are deployed and managed. The system optimizes critical operational factors, including response latency, request throughput, and power consumption. By maximizing the efficiency of every cycle, the platform helps enterprises extract significantly more value from their existing GPU investments.

Headquartered in Bengaluru with a growing global footprint, the startup offers organizations flexible deployment options. Companies can run AI models through Simplismart’s managed platform or deploy the software directly within their existing cloud environments and on-premises data centers. In addition to deployment capabilities, the platform furnishes tools for fine-tuning models, automatically scaling computing resources in response to fluctuating demand, and monitoring real-time system performance. Furthermore, Simplismart enables cloud providers to build and operate their own dedicated AI inference services on top of raw GPU infrastructure using its specialized software stack.

The market validation for this approach is reflected in Simplismart’s expanding enterprise client roster. The startup reports that its software helped AI video-generation platform InVideo drastically slash its GPU serving costs by 56 percent. Simplismart also provides technological infrastructure for other prominent players in the generative AI space, including video platforms HeyGen and Higgsfield, as well as enterprise AI startup Ema. Additionally, Indian cloud computing provider Yotta utilizes Simplismart’s technology to power the serverless AI inference service embedded within its Shakti Studio platform.

This technological momentum has translated into substantial financial growth. Jain revealed that Simplismart’s annualized revenue run rate has surged 14-fold over the preceding six months, reaching an eight-figure sum. This metric is calculated using recent customer usage patterns rather than strict contracted recurring revenue, and it explicitly excludes pass-through expenses paid directly to underlying infrastructure providers, according to the company. Jain declined to disclose the exact revenue figures. Notably, more than half of Simplismart’s total revenue originates from the United States, even though the startup maintains a larger volume of individual customers in its home market of India.

Beyond financial backing, Nvidia’s involvement extends into deep technical collaboration. Jain noted that Nvidia has been working closely with Simplismart on several technical integrations, particularly concerning Nvidia’s Dynamo inference software. This partnership sees Simplismart’s engineering team collaborating directly with Nvidia’s solutions architects. Furthermore, Nvidia has actively assisted the startup by facilitating introductions to prospective enterprise customers and industry partners.

This strategic partnership introduces an interesting paradox: Nvidia, the undisputed dominant supplier of advanced AI chips, is actively investing in optimization software that can theoretically allow enterprises to run specific workloads with fewer GPUs. Addressing this dynamic, Jain argued that making AI inference more efficient does not cannibalize hardware sales. Instead, by lowering the total cost of ownership and easing operational friction, improved efficiency ultimately benefits Nvidia by making its hardware ecosystem more economically attractive to a broader base of customers, thereby encouraging wider, more sustainable adoption of AI applications.

Simplismart operates within an increasingly crowded and competitive marketplace for AI inference software. The startup competes directly with native offerings from Nvidia itself, as well as massive cloud computing hyperscalers including Amazon Web Services, Google Cloud, and Microsoft Azure. A rapidly expanding cohort of specialized startups is also developing tools aimed at accelerating AI models and reducing operational costs, propelled by explosive global demand for AI applications and the acute economic necessity of maximizing expensive GPU resources.

"Where we tend to win is where people have to be more stringent on their SLAs and TCO," Jain explained, referencing demanding service-level agreements and tight corporate cost constraints. Unlike rival solutions that offer rigid, preset options geared purely toward maximum speed or minimal cost, Simplismart focuses on striking a tailored balance suited to each individual customer’s operational parameters.

Simplismart was founded in 2022 by Jain, a former Oracle engineer, and Devansh Ghatak, a former Google engineer. The co-founders first met while studying at BITS Pilani, drawing upon their combined academic background and industry experience in building and deploying complex machine-learning systems. The startup currently employs a workforce of 58 people.

With the closure of its latest funding round, Simplismart plans to allocate capital toward expanding its operational presence in the United States, growing its engineering and go-to-market teams, and accelerating the research and development of its core inference software. Jain emphasized that the U.S. remains a vital geographic market as the company seeks to build upon its established customer base there.

The funding announcement follows recent media reports by Indian financial daily The Economic Times suggesting that Simplismart had engaged in preliminary discussions to raise a significantly larger funding round at a substantially higher valuation. Asked about those reports, Jain declined to comment, maintaining that the startup was not actively engaged in fundraising and confirming that the current Series B round is fully closed and will not be extended.