Article complete

Can Simulated Users Replace Weeks of Testing Products? Vectorial AI Thinks So

The startup is targeting businesses in India, the US, and Canada with AI-generated audience simulations. It plans to expand its behavioural models and product offering.

consumer behaviour
Image Credit - Nalini Nirad

Key takeaways

  • Vectorial AI aims to streamline consumer research by simulating audience behaviour with its SAPIENS model.
  • The startup has secured $3 million in seed funding to enhance its behavioural models and expand its product offerings.
  • Founded in 2024, Vectorial is headquartered in San Francisco with a team based in the US and India.
  • The cofounders have extensive backgrounds in modelling human behaviour and AI research from prestigious institutions.

Takeaways drafted with AI from the article text.

As AI tools and products move from pilots to production, the bottleneck in software development has shifted to maintaining production quality. According to Google’s DORA report, higher AI adoption correlates with faster delivery, but it is bogged down by inconsistent outputs that strain system stability and require debugging. 

The need is now to accelerate testing by filtering out weak prototypes and evaluating features against modelled users. 

Vectorial simulates such consumer behaviour with AI. Its SAPIENS model creates virtual representations of target audiences so product, design, and marketing teams can test ideas before launch.

“We are providing the most active population models that are made from real behaviour of people of the company's target customer,” Co-founder and CEO Taranveer Singh tells AIM.

The startup was founded in 2024 and is headquartered in San Francisco, with a 12-member team across the US and India. It raised $3 million in seed funding in October this year from SenseAI Ventures, Smilegate, and investors associated with the Berkeley and Stanford ecosystems.

Modelling Human Behaviour

“I've spent the last 10 years modelling human behaviour,” says Singh, who started his career in defence research, worked with Andrew Ng's team at Coursera while at Carnegie Mellon University and later joined a research lab that predicted career decisions.

Before joining Berkeley AI Research as a visiting research fellow, Singh began working on SAPIENS and benchmarked it against other behaviour models—which led to Vectorial. He met his co-founders around the same time.

Jasmine Kaur, the chief business officer, graduated from Stanford Graduate School of Business and worked on synthetic users for marketing at Pfizer. Rahul Garg, the chief product officer, has worked in product leadership at Blinkit and Housing.com and on a behaviour graph for sales conversations at SaaS platform Zime AI.

The founders began exploring commercial uses in 2024, worked with design partners from mid-2025 and launched in January 2026. SAPIENS began as a question-and-answer interface that worked like a focus group with a simulated audience and now lets businesses test software interfaces, physical product concepts, and advertising campaigns.

Change User Research

Singh believes a traditional study, from recruiting participants to analysing findings, can take three to 12 weeks, depending on scope. AI has sped up product development, but “user research becomes a bottleneck,” he says. Some companies skip research because of time or cost, which he said can waste millions of dollars on failed products.

We are providing the most active population models that are made from real behaviour of people of the company's target customer

Taranveer Singh

SAPIENS is built on behavioural data from partners and AI-moderated interviews with paid participants, which the company uses to profile how people behave across locations, lifestyles, and income groups.

Vectorial says it has modelled more than 2.5 million users, including 500,000 in India, and run more than 30,000 simulations.

The company aims to deliver feedback within hours at around one-twentieth of the cost of a conventional study. Vectorial charges a base platform fee to develop an audience model, then runs additional simulations.

The company argues that simulations can replace conventional research for many decisions, though it still encourages calibration against live participants. “During the calibration phase, we encourage them to do these studies with live people so that we can show how accurate we are,” the co-founder says.

Can Simulated Users Predict Real Behaviour?

With quick testing and simulations comes the challenge of reliable accuracy. “It's one thing to have a model accurate sometimes for some audiences, but it's another thing to have a model accurate in all possible scenarios that a customer can influence,” Singh notes.

Vectorial addresses this with a broader pipeline of behavioural data. To build enterprise trust, it runs a calibration period that compares simulated responses against previous studies and interviews with real people, and the model gathers more data where it finds gaps.

The company reports that SAPIENS is 87% accurate in predicting behaviour and has matched previous customer research by 85–95% in some comparisons.

Researchers remain cautious, with a 2026 survey by User Interviews finding that 47% were sceptical and wanted more evidence and 88% cited quality and accuracy of insights as a concern.

“We are not making up users,” he insists. “These are like real people who we are modelling against and real behavioural data.”

SenseAI Ventures and the Road Ahead

Vectorial’s announced customer list includes boAt, GoodRx, Suki AI, MIDI Health, and Concrete Software. It targets healthcare, gaming, and technology companies building software products and digital experiences, and has not disclosed its current revenue or run rate.

Its competitors include Maze and UserTesting, along with AI-moderated research companies Listen Labs and Outset AI. Singh notes that AI-moderated interviews can accelerate research but are limited to about 100 people, and each new question requires a new study. Vectorial instead keeps a model of the audience that customers can query repeatedly.

On what drew SenseAI Ventures to Vectorial, Singh adds, “When we first pitched them, they clearly understood that we are a category creator in this, and behaviour models and simulations are going to be the new way to build and launch products.”

The company will use the seed funding to develop SAPIENS 2.0, expand behavioural data partnerships and invest in product-led growth. Its tools include a Chrome Web Store beta and a Figma plugin, with an MCP server planned so compatible AI tools can interact with simulated audiences.

Vectorial adds around five customers per quarter and expects that to accelerate, with a lower-cost product for individual developers planned. The startup is targeting double-digit million-dollar revenue by the end of 2027.

Published

Spotted an error? Tell our editors. Read our editorial standards and corrections.

More in this story

AI in India · 5 stories
  1. India's AI Funding Boom in H1 2026 Left Most Startups Starving
  2. Meet the AI Startup Behind India’s Got Latent Viral Song ‘Chai Ki Tapri’
  3. Most of India's Nearly $500 Mn AI Boom is Squeezed Into Just 4 Deals
  4. The Hardest Test for India’s AI Dream
  5. You are here: Can Simulated Users Replace Weeks of Testing Products? Vectorial AI Thinks So
About the author

Journalist

Ridhika is a journalist covering the real-world impact of artificial intelligence at the intersection of tech, policy, and society.

All stories
Belamy · Daily

The day’s AI news, before your first meeting.