PanScience Innovations
← Media & insights
Behind the BuildMay 1, 2023

Behind the Build: Three Years of PredCo, One Factory Floor at a Time

Bhavay Garg | Co-founder, PredCo
Behind the Build: Three Years of PredCo, One Factory Floor at a Time

Bhavay Garg, Co-founder, PredCo

We started PredCo in May 2023, and it still feels like we began yesterday. The idea was easy to say and hard to do: help factories know when a machine is about to fail, before it actually does. Three years later, PredCo has grown into something much bigger than that first idea. Here is how it happened.

It started with one idea

In 2023, predictive maintenance was still a new idea in India, though it was already taking off in the US. Anshul sir saw the opportunity early and wanted to build something around it. When my classmates told me about it after class, I was curious enough to meet him and Dheerendra sir. By the end of that first conversation, I knew I wanted to be part of it.

Why I said yes

There were two reasons. First, I had always wanted to start something of my own. Here was a chance to do it with people who had already built big companies, which meant a mentor who could tell us what to do and, just as importantly, what not to do.

Second, the space itself. I have loved hardware, robots and sensors since school, and PredCo sat right in that world.

From data pipelines to predicting the future

We set out to build a product, and it began with the basics. Our first job was to build the sensor system that collects data from real machines and feeds it into our first data pipelines. From there, we built analytics on top of that data, with a bigger aim in mind: to predict what could go wrong before it actually does.

Dheerendra sir guided us closely through those early days. We met almost every day to talk through how the platform should come together and which tools to use. The moment predictions first came out of the data from our sensors on the ground felt like our first real win. It sounds small now, but it felt huge then.

The platform has come a long way since then. We started with something that looked and worked very basic, and today it is a professional product. We built and refined its features through conversations with many customers, and those conversations also shaped our roadmap.

AI is the next layer. It works on the data the platform collects, helping us spot patterns and predict what is likely to fail.

Reaching our first clients

Towards the end of 2023 we started taking PredCo to exhibitions. Our first conversations were with a few tractor and auto companies, and that is how we began reaching clients.

What PredCo does today

We began with predictive maintenance, but the more clients we spoke to, in India and abroad, the more problems we found worth solving. Today PredCo also works on compliance for manufacturing and similar industries, on vision-based monitoring of day-today processes, and on using AI to improve how these industries run.

Seeing how things are made

We have spoken to hundreds of clients and visited around 40 to 50 plants. This is the part of the job I enjoy most.

Every visit introduces us to something new. Whether it is understanding how steel is manufactured, how oil reaches a petrol pump or how energy is generated and transmitted, these experiences give us a completely different perspective on industries we often take for granted.

Take oil. It is normal to pull up at a petrol pump and fill the tank, but very few people see how that fuel gets there, from refining to delivery. Energy is a sector we work in a lot, and learning how power is generated and transmitted never gets old. Every new client opens a new window.

The hardest part

The biggest challenge has stayed the same since day one. The industries we want to work with are very traditional, and many are slow to adopt new technology. In many places basic logging is still done on paper, and some have no proper ERP system. Then we walk in and talk about AI.

In my experience, manufacturing in India is still early in its digital journey, with automotive as the exception.

Looking Beyond India

There is definitely scope abroad, but also competition. Many companies in Europe and the US either have in-house teams or already buy from someone, especially for the basics like predictive maintenance, process monitoring and worker monitoring. We have still found success with a few completely new use cases the market didn't have.

What I would do differently

If I started PredCo again today, I would talk to clients very differently. When we began, our outreach was generic and we didn't know how clients think. Now we know which topics land. Compliance, for example, gets a much better response.

Positioning and pricing are the other big lessons. We have lost deals on price, and those losses taught us a lot. I already brought the technical side with me, but knowing how to position a product, price it and talk to customers is what would change the most.

Life at PSI

What I value most about PSI is its energy. Things get built here at a pace I have rarely seen elsewhere, decisions are made quickly, and you can reach anyone when you need a resource or an answer. It keeps you on your toes, and after three years it has become second nature.

PSI also adds credibility and depth to PredCo. Its ventures work with the judiciary, including the Supreme Court, and with hospitals and large healthcare chains, and those names carry weight with clients. With ventures building for very different industries, domain expertise is always close at hand, so a conversation is usually all it takes when we feel stuck.

Looking ahead

If there is one thing PredCo has taught me, it is that building a startup rarely follows the path you first imagine. Your understanding of the problem keeps changing as you speak to customers, visit plants and face challenges firsthand.

What started as an interest in hardware and predictive maintenance has grown into a much broader learning experience. Three years in, PredCo is still a work in progress, and there is always another problem to understand and another opportunity to build something better. That is what keeps it interesting.

For me, the most exciting part is that there is always something new to learn, another problem to understand and another opportunity to build something better.

More from Behind the Build
November 3, 2026

Beyond the Code: A Journey of Learning and Growth at PanScience

Read →
October 12, 2026

Growing Beyond the Ledger: My First Two Weeks at PSI

Read →
August 3, 2026

Exploring Beyond the Classroom

Read →
← Back to all media