With over 10 years of experience in the technology field, I am a passionate and innovative leader who strives to create positive social and environmental impact with IoT, clean-tech, and AI solutions. As the CTO of Oizom, a company that develops smart instruments to sense and analyze natural resources, I lead the vision, strategy, and execution of cutting-edge products that help monitor, visualize, and predict environmental data. I am driven by the mission of becoming an environmental big data company by 2025, and leveraging the power of cloud software and AI to empower stakeholders, policymakers, and citizens to make informed decisions for a better planet.

In my current role, I leverage my skills and expertise in robotics, wireless networking, and computer vision to design, develop, and deploy IoT devices that sense and measure air quality, water quality, soil health, and energy consumption. I also oversee the data science and machine learning team that analyzes and interprets the data collected by the devices, and generates insights, reports, notifications, and predictions using AI. I have co-authored a book on computer vision, and have published and presented my work at various international conferences and competitions. I am also a co-founder of Jayso Labs, a not-for-profit organization that helps students, hobbyists, and developers who want to explore electronics technology: I am always eager to learn new things, collaborate with others, and challenge myself to create innovative and impactful solutions.

🚀 We benchmarked Sarvam AI vs Gemini for Gujarati Newspaper OCR — here’s what we learned.

At TownPlanMap, our agent crawls Gujarati newspapers every morning to detect real estate public notices.

From each notice, we:
• Extract survey numbers
• Identify owner details
• Understand legal context
• Map notices to geolocations

This is not “just OCR.”
It’s structured extraction from dense, legal, vernacular documents.

We tested both Gemini and Sarvam AI Vision on real Gujarati public notices from production workflows.

While both models performed well, we observed that Sarvam handled Gujarati legal formatting and vernacular nuances more reliably — especially in structured elements like survey numbers and contextual clarity.

In document intelligence pipelines, small improvements in extraction quality make a big difference downstream.

It’s great to see focused innovation happening for Indian languages.
Kudos to the team at Sarvam AI  and the Vivek Raghavan and Pratyush Kumar for building serious infrastructure for vernacular AI 🇮🇳

Sharing some comparison visuals in the post 👇

#AI #OCR #Gujarati #VernacularAI #DocumentAI #PropTech #TownPlanMap

Sohil Patel

Why the Maker Spirit Still Matters
There’s a different kind of energy inside a makerspace.
Less talk.

More building.

Real problems. Real constraints.
Recently, I spent time with a student team in Ahmedabad preparing for the FIRST Tech Challenge. We went deep into their vision system, especially around AprilTag detection and how it behaves outside controlled lab conditions.
What stood out was not just the tech.
Alongside building their robot, the team has created a community platform to help other students competing in the same challenge. A shared resource hub built using vibe coding, focused on learning, not just winning.
That mindset matters.
I’m glad to support this initiative through Jayso Foundation, because this is how strong engineering ecosystems are built. Not just by shipping products, but by enabling others to build.
The future of Indian R&D is being shaped in rooms like these.
The maker spirit is alive.

And it’s worth backing.

Had a great time in a panel discussion at eChai Ventures focused on the realities of building and scaling hardware manufacturing in India, alongside Rutvij Dasadia and Er. Nandan Shukla, hosted by Shruti Jayswal. 

The discussion stayed refreshingly practical how manufacturing actually works in India, where it breaks, and what you only learn once you start building at scale.

From Oizom’s side, the conversation focused on what it takes to manufacture environmental sensing hardware reliably in India. We spoke about designing products with manufacturability in mind from day one, managing sensor variability across batches, building calibration and quality systems that scale, and why field feedback from real deployments must continuously flow back into R&D and production.

Manufacturing at scale isn’t just about cost or capacity. It’s about process discipline, traceability, and building systems that behave consistently across climates, geographies, and years of operation.

Huge appreciation to Jatin Chaudhary and the eChai Ventures team for consistently creating spaces where founders can have honest, experience-led conversations.

Here it is:
Just walked out of the India AI Impact Summit 2026, Delhi.
One moment stayed with me the whole flight back.
A 16 year old from Pune was presenting an AI model that predicts crop failure before it happens. No funding. No mentor. Just public datasets and curiosity.
Honestly, that hit different.
We keep telling young India to wait. To qualify. To be ready.
The summit said something else entirely.
2 Million new jobs coming by 2030. 22 languages finally understood by machines. Kids between 13 and 21 solving problems that stumped experts.
India is not consuming technology anymore. We are building it. For us. In our languages. For our problems.
Swipe through. I broke down 6 real takeaways from the summit that actually matter for your career.
Which domain are you most excited about? Health, Agri, Climate or Safety? Tell me below.

taplio

Sohil Patel's Linkedin Analytics

Sohil Patel

Performance ScoreSohil Patel's LinkedIn Performance

What is Sohil talking about?

Sohil Patel's Best Posts (last 30 days)

Ready to grow your LinkedIn presence?

taplio

.css-6ytdur{background-image:linear-gradient(284deg, rgba(39, 43, 72, 1) 0%, rgba(33, 47, 183, 1) 100%);color:transparent;-webkit-background-clip:text;background-clip:text;}Sohil Patel's Linkedin Analytics

Sohil Patel

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Sohil Patel's LinkedIn Performance

What is Sohil talking about?

Sohil Patel's Best Posts .css-1emoo84{font-size:var(--chakra-fontSizes-md);font-weight:400;color:var(--chakra-colors-gray-600);}(last 30 days)

Ready to grow your LinkedIn presence?

Sohil Patel's Linkedin Analytics

Performance ScoreSohil Patel's LinkedIn Performance

Sohil Patel's Best Posts (last 30 days)