Vishnu Vasanth and the cost of asking one more question

A business can become better informed and poorer at the same time. Every new dashboard, customer interaction and question has to be processed somewhere. Vishnu Vasanth has built a company around the awkward moment when enthusiasm for data meets the invoice. The information may be valuable. Finding out what it means can become an expensive habit.

Vasanth is the co-founder and CEO of e6data, which develops a compute engine for demanding analytics and AI workloads. His pitch begins with a concession that enterprise buyers might appreciate: they have already made substantial investments in their data systems. Asking them to relocate everything before helping them is a peculiar opening offer. He wants to bring a different engine to the data they already have.

That argument sits at the intersection of his working lives. He has been an engineer, an investor and a renewable-energy entrepreneur. His career has moved between the machinery that makes something possible and the arithmetic that makes it worth doing. At e6data, both questions arrive on the same bill.

Before the query engine, a different kind of power

Vasanth studied electronics and telecommunications engineering at RV College of Engineering and finance and strategy at the Indian School of Business. The pairing is useful background for a founder selling a technical product through an economic argument. One qualification concerns how systems work; the other concerns how organizations decide what to pay for them.

His early career included technical marketing at National Instruments, followed by business development and technical marketing at Agilent Technologies. He subsequently worked on the investment team at PremjiInvest. These roles placed him in different parts of the relationship between technology and capital: explaining a product, developing business for it, and evaluating businesses as investments.

Then came RenewGEN, where he was co-founder and CEO. His public founder profile describes that chapter as cleantech entrepreneurship with an exit. Renewable energy and database software make an unusual pair on a résumé. Both, however, put infrastructure at the center of a business. A useful output depends on the system underneath it, and the cost of running that system matters.

He later became director of investments at Equicap Asia, before taking the co-founder and CEO role at e6data in March 2020. The sequence resists the tidy version of a founder biography in which every previous job appears to have been an audition for the last one. There were businesses to build and investments to assess before there was a query engine to sell.

The school register hiding inside the lakehouse

One of the more revealing pieces of Vasanth’s public work starts in a classroom. In a November 2024 essay co-written with Karthic Rao, he explains modern data infrastructure through the attendance ledger: names down one side, dates across the top, a teacher recording who turned up. Before the technical vocabulary arrives, there is a table anyone can recognize.

The example gives a complicated subject a human scale. A single teacher updating a register has a manageable task. Many people reading and changing information at once need more elaborate rules. The essay follows that progression through databases and warehouses to object storage, where large collections of files can be kept separately from the engines used to analyze them.

Open table formats such as Apache Iceberg, Delta Lake and Apache Hudi help organize those files into usable tables. The distinction matters to Vasanth’s business: rules for storing and updating information leave room for different engines to process it. The attendance register has acquired a considerable supporting cast. Its basic purpose remains legible.

For a founder working in a field thick with acronyms, choosing a school ledger is an effective act of translation. It lets readers follow the reason for the technology before asking them to remember its name.

A funding round with an argument attached

In September 2024, e6data announced a $10 million Series A led by Accel, with participation from Beenext and others. Vasanth’s announcement concentrated on the workloads that customers found particularly difficult and expensive. He also objected to forms of compute lock-in that narrowed customers’ choices even when the wider data architecture was described as open.

That is a more specific complaint than a general request for cheaper software. An organization might choose a data format freely and still find that switching the engine used to analyze it is painful. Vasanth wanted e6data to work across cloud providers, table formats and catalog systems, so that selecting a processing engine could become a practical decision.

Accel had first invested at seed stage in 2023. Its account of the Series A described early design partners reporting 50 to 70 percent lower overall costs and query speed improvements of five to ten times. Those figures describe reported customer experiences in that announcement. They give the proposition a scale without turning it into a promise for every workload.

At the time of the round, the company said the money would support research and development, customer onboarding and hiring. It was expanding a design-partner program with production support and professional services. A query engine can be admired in a demonstration; getting it into a customer’s working environment involves another set of obligations.

The persuasive power of keeping the furniture

In explaining the business in 2024, Vasanth described a specialist engine for difficult enterprise workloads that could run alongside an organization’s existing platforms. The proposition included the chance to test it in shadow form before committing to a switch. A purchasing decision could begin with a comparison on the customer’s own work.

There is a recognizable courtesy in that approach. An enterprise data system represents accumulated decisions, habits and dependencies. Its owners may want a faster answer without wanting a new home for every table. Keeping the furniture makes the discussion less theatrical, and potentially more useful.

“We offer something that is tailored specifically for high-end enterprise workloads.”

Vishnu Vasanth · September 2024

The current product documentation makes this concern with coexistence concrete. It describes connections for familiar tools and protocols, alongside catalogs, access policies and cluster operations. Those details rarely supply the opening line of a founder’s pitch. They help determine whether a team can use the product after the meeting ends.

Vasanth’s subject also extends beyond a single sales conversation. In April 2024, he joined Joe Reis for a podcast discussion covering analytical query engines, shifting left and the Indian technology scene. The published episode runs just over 52 minutes. This corner of computing rewards a willingness to stay with the explanation.

A room full of people asking better questions

His public appearances offer another view of the work. At Big Data & AI World Asia in Singapore, he spoke with Ravit Jain about the tension between rising analytics costs and the speed and concurrency customers wanted. In his own account of the conversation, he emphasized customers, partners and the wider data community.

That community has also gathered closer to e6data’s engineering base. A Lakehouse Days event at Accel LaunchPad in Bengaluru listed Vasanth alongside speakers from Onehouse and PayPal, with a program focused on open table formats. The invitation described talks, workshops and panel discussions. The same architectural choices being debated in product meetings could be discussed around a shared technical problem.

At the AWS AI Conclave, Vasanth and Adishesh Kishore presented SQL and vector-search capabilities, with Kishore demonstrating a search against a reviews table. It was a concrete way into a broad AI conversation: begin with information a business might actually hold, then show how someone could ask a different kind of question about it.

When the next question comes from software

By 2025, Vasanth’s public agenda included an AI Council session titled Everything Everywhere All at Once: Object Store Native. The cinematic title introduced a technical thesis about more kinds of applications being built on open storage foundations. Analytics, AI and other workloads would place fresh demands on the engines underneath them.

At MachineCon USA that year, he described a shift from a relatively small group of executives consulting delayed information toward many more workers expecting current answers. He also pointed to software agents as future consumers of compute. In his account, greater demand, fresher data and harder questions compound the pressure. That is his case for pursuing efficiency as AI use expands.

His 2026 conference listing continued in that direction. Data Decoded Manchester scheduled him for an October 14 session on agent traces and proprietary intelligence. The vocabulary has moved beyond the query-engine conversations of earlier appearances, but the recurring concern is recognizable: the intelligence an organization creates should remain useful to that organization.

Vasanth’s career supplies an unusually varied route into this argument. Technical marketing, investment and renewable-energy entrepreneurship each precede the present business. Now he is asking enterprises to examine the economics of their curiosity. There will always be another question to ask of the data. His wager is that answering it should become easier to justify.