Exclusive: How Hyderabad startup Vatins is reinventing AI with privacy-first sovereign intelligence
HYDERABAD: In a world where Artificial Intelligence(AI) has become synonymous with feeding data to powerful servers owned by tech giants, a bootstrapped startup from Hyderabad’s Financial District is building something deliberately different, AI that never touches the internet. AI that stays inside your organisation, learns from your data, and never leaves.
This is the philosophy behind Vatins Systems Pvt. Ltd., founded by Sunny NV. With 19 years of experience spanning cybersecurity, digital forensics, and cyber investigations, Sunny has worked extensively with India’s premier investigation and law enforcement agencies, while also helping enterprises across diverse industries strengthen their cybersecurity posture. As a result, governments, police forces, and businesses that need to keep data secure are turning to this approach.
In an exclusive conversation with HyderabadMail Sunny NV said, “The most valuable data should never have to leave an organisation’s walls to become intelligent.” Sunny says this belief sets Vatins apart from most companies that claim to use AI. He explains that many firms just connect to commercial AI tools, send them data, build a user interface, and call it a product. In contrast, Vatins takes a different path. They use open-source language models, customise them with each organisation’s private data, and keep everything inside the company’s own network. The AI never connects to external servers.
He calls this approach Sovereign AI. He believes only Sovereign AI can truly protect sensitive data for governments, law enforcement, and regulated companies. “Imagine an organisation with 25 years of accumulated data, for instance law enforcement agencies with classified records. If that data is fed into ChatGPT or Gemini, and those servers are ever compromised, everything is exposed. Why not use AI securely? Keep it inside,” said Sunny.
Journey from Cybersecurity to Entrepreneurship
Sunny did not come from the startup world. Instead, he built his career in intelligence and cybersecurity. He began his consulting journey in 2008 and has since worked with the Indian Defence and multiple State Police and Law Enforcement agencies across Counter Terrorism, Counter Narcotics, Cyber Crime Investigations, Women & Child Protection, and Digital Intelligence. His focus was always cybersecurity for national infrastructure and security.
This unique background gave him insights that later shaped Vatins. From those years in Premier investigation agencies, he brought a specific understanding of how crime data is actually generated and stored, fragmented across dozens of systems, inaccessible in practice to investigators who need it most. That understanding became Vatins’ founding insight.
Vatins, which began operations in 2015. The company has stayed bootstrapped until now, without outside funding or venture capital. It has grown for more than a decade through quality engineering and referrals. Some international clients have worked with Vatins for over eight years.
Explaining his startup operations Sunny says, “We have never been in a comfort zone. That struggle made us resilient. About 90% of startups fail. We want to build this independently. When the time is right, we might seek funding, but not yet.”
IntraGPT: AI that works entirely offline
Vatins’ products tackle the problems Sunny spent decades studying. The flagship is IntraGPT, a Sovereign AI platform based on Closed Source Intelligence (CSINT) principles. IntraGPT runs entirely within a police or government network. Investigators can ask questions in plain language, just like with ChatGPT, but all data stays inside the organisation.
In addition to IntraGPT, Vatins has developed other AI tools for law enforcement. Sunny explained, “Imagine a new officer joining today who knows nothing about a case from twenty years ago.They type a query in natural language and get an answer. We can detect patterns, uncover hidden connections, and make predictions, all from data that was previously sitting in silos.”
Vatins also developed CSight, India’s first Sovereign AI tool for handling suspected Child Sexual Abuse Material (CSAM) entirely offline. Normally, officers must watch hundreds of videos and images, which can traumatise them. CSight scans the material automatically, identifies CSAM content, and gives officers timestamped findings, without showing them the full material.
Sunny says, “We want to reduce trauma and speed up analysis. Officers no longer need to watch hour-long videos. You upload a hundred files. The system tells you pin point, for example, ‘At 2 minutes, 13 seconds, this file contains CSAM.”
Vatins also offers RedPanda AI, which monitors the surface, deep, and dark web for threats. LAPIS is an AI-powered system that automates legal and policing workflows. CDAT AI analyses call data records and tower dumps. Project Integrity helps the country’s Anti-Corruption Bureau with AI-assisted investigations.
Beyond its AI platforms, Vatins serves as the cybersecurity consultant to the governments of both Telangana and Andhra Pradesh, contributing to the protection of digital ecosystems that serve more than 60 million citizens combined.
The company has also delivered solutions internationally, with clients in the United States, Southeast Asia (including a long-standing eight-year engagement in Malaysia), and the Middle East. Sunny added among its client portfolio are Gartner-recognised organisations, a benchmark that reflects the calibre of enterprises the company has been trusted to protect.
Recently, Vatins opened an office in the United States, aiming to expand its AI data intelligence solutions to more global clients. “Americans invented the foundational technology, Open AI, Gemini, but the utilisation of that technology at the ground level, especially for investigative and intelligence purposes, has not yet reached its potential. There is a huge opportunity for us there,” said Sunny.
Ask Sunny what is most misunderstood about AI from a cybersecurity perspective, and his answer is immediate, its dual-use nature. He points to Anthropic’s Claude Mythos model, released to detect security vulnerabilities, as an example of AI’s extraordinary offensive and defensive potential. The model reportedly identified a vulnerability in OpenBSD that had gone undetected for 27 years, despite decades of code reviews by cybersecurity experts and automated security tools. The same capability that makes such a model invaluable for defenders also makes it extraordinarily dangerous if it reaches the wrong hands.
Cybersecurity risks and opportunities of AI
According to Sunny, “It is like a gun. The same gun is with the police. The same gun is with the anti-social element. The question is only how it is used.” His team stays current by participating in bug bounty and responsible disclosure programmes run by major technology companies, an approach that keeps Vatins’ engineers at the frontier of offensive security. Team members have earned bounties through responsible disclosure programmes at companies including Nvidia, Impact Guru, and Yahoo.
For young graduates considering a career in cybersecurity, Sunny’s advice is practical and specific. Start internships from the first or second year of graduation, not the final year. Register on platforms like Internshala and build a visible portfolio of practical work. Participate in Capture the Flag (CTF) competitions and bug bounty programmes to develop real-world offensive security skills that academic curricula rarely teach. And follow platforms like TryHackMe to simulate professional security testing environments before entering the field. “By the time an employer comes to you, you need to be in a position to highlight what you have actually done, not just what you studied,” he says.
For entrepreneurs entering the technology space, his advice is equally direct, build solutions, not just services. Services generate revenue but hold you in place. Solutions create compounding value and open doors that service delivery alone cannot.
“We spent years delivering services. Now, we build solutions, and that shift has changed everything,” he says. Hyderabad based Vatins Systems, after a decade of steady growth, the company has reached a turning point. Vatins has a strong record in sovereign AI and cybersecurity consulting, a growing international presence, and a vision to deploy AI in at least 25 Indian states to help police automate investigation work.
The company will stick to its founding philosophy as it grows. Sunny concludes, “The future of AI will be shaped by trust, not size. The best systems protect privacy, support national sovereignty, and help organisations unlock knowledge from their own data. Bring AI to the data, never the data to the AI.”
