Imagine if Siri or Alexa went to school, got an advanced degree in Petroleum Engineering and Geoscience, then went to work in the oilfield for several years, attended technical conferences, read journals and books; envision the product and the process, and you’d have NESH, the Smart Assistant for oil and gas. In addition, you have a chance to try out NESH and collaborate and explore a customizable artificial intelligence. It’s a tremendous opportunity. Welcome to an interview with Sidd Gupta, founder of NESH.

What is your name and your background?

My name is Sidd Gupta and I am a Petroleum Engineer by education. I have an undergraduate degree in Petroleum Engineering from Indian School of Mines and a Master’s Degree in Petroleum Engineering from UT, Austin. I have worked in the oil and gas industry for a little over a decade now with roles involving working at an offshore well site with the wireline crew and interpreting open hole and cased hole logs, working on an oil platform to ensure operational safety, providing petroleum engineering consulting to large and small E&P organizations, operating in resource plays and to managing the development and growth of two of the largest Petroleum Engineering Software. I am a Schlumberger and Shell alum who transitioned from enterprise oilfield leaders to starting my own Energy Technology Startup.

How did you get interested in analytics?

When I was younger, I was a fan of Arthur Conan Doyle’s Sherlock Holmes books and, of late, I have binged the BBC series too. One of the things I noticed as I read these stories and watched the show was that one key advantage that Holmes enjoyed wasn’t more data than anyone else, but his ability to see that data differently; finding connections and patterns in seemingly unexpected and regular places. I wanted to solve mysteries and see patterns like him, but unfortunately, grad school and the job didn’t take me to the seedy alleyways of London, so I went about my life in a relatively uneventful city of Houston. Then things took a turn in 2014. Around November, when the oil prices started falling, I was intrigued to know everything about it. This was the first major downturn of my oil and gas career so I was determined to emerge wiser from it. Why was it happening? What factors were responsible for it? Can it be predicted? What other things were affected by it? The quest to answer these questions got me interested in analytics and looking deeper into data. For me, analytics, data science, machine learning, advanced statistics has always been a means to an end - To learn something we didn’t know before and communicate that effectively to others.

What is your latest project?

As it happened, that during the 2014-2016 downturn, one of my friends lost his job as a Reservoir Engineer at an upstream Oil and Gas company. He tried applying to many companies but never got an interview. An acquaintance of mine who worked at an E&P company was looking to hire a Reservoir Engineer for their onshore development team, so I connected them to see if there was any synergy. Long story short, he didn’t get hired but what stumped me was the reason for not hiring him. It wasn’t due to a lack of technical understanding, or experience, or an unpleasant interview but because he didn’t know how to use the reservoir simulator that was used in the company. In his previous job, he had been trained in a different software. He has since found a new job but that specific incident stayed with me and as I thought more about it, I realized, that this is a systemic thing in our industry. Almost every new job post has a requirement - ‘Must know how to use XYZ software’. And there is a good reason behind it. Oilfield science is complex and the software that we have developed to leverage that science is also complex, so companies spent a lot of money on training their staff how to use the software and going forward they prefer to hire people who already know how to. Makes sense. However, with the increased activity in Shale plays, there has been a correspondingly huge rise in the volume, variety, velocity, and veracity of data. And this new data complexity and the existing software complexity don’t mix well with each other. So we decided to solve this problem. Analyzing the data to derive meaningful and valuable knowledge was half the battle. The other half and perhaps the more important half was to make this knowledge effortlessly accessible. That is the project we are currently working on and that’s how we came up with the idea of Nesh.

What is Nesh and what can it do?

Nesh is the Smart Assistant for oil and gas. She helps oil and gas companies make better and faster decisions, by organizing the collective knowledge within the company and making it effortlessly accessible. Imagine if Siri or Alexa went to school, got an advanced degree in Petroleum Engineering and Geoscience, then went to work in the oilfield for several years, attended technical conferences, read journals and books; you would have Nesh.

Nesh runs on any browser. Users can simply ask, and she will answer their complex technical questions, in a conversational way. Nesh is designed to run powerful petroleum engineering and geoscience workflows behind a simple conversational UI. Nesh does away with the complexity of traditional Oilfield software by combining high-fidelity science with Natural Language Processing and other Machine Learning Techniques to create a simple, interactive and enjoyable experience.

How is it different or similar to Alexa?

It is similar to Alexa in that both are conversational agents that can provide answers. But the similarities end there. Unlike Alexa, Nesh is a device independent smart assistant that runs on any browser and has been custom built just for upstream Oil and Gas. You can either talk or chat with Nesh. With Alexa or Siri, when things don’t go as expected (“I am sorry I am having trouble understanding you right now”), you cannot engage with Amazon or Apple to provide feedback and help these assistants get better, but with Nesh you can take control and can train her the way you like to understand more phrases, more questions, and more workflows so she can getter smarter over time with the help of your Petrotechnical expertise.

What makes Nesh unique?

When we started building Nesh, we reimagined oilfield software from the ground up. There were four main areas we were hyper-focused on - Ease of Access, Connection to disparate data types, analysis of data using domain-centric and analytical methods and an open architecture.

Using Nesh, companies can organize their data (structured and unstructured), run domain-centric analysis on it and get answers effortlessly. We built Nesh using a combination of deep oilfield science (backend) with a simple conversational interface (frontend) to give an Oil and Gas company a tool they can use without any friction. While simplicity is important, the combination of deep science is critical because when drilling or producing a multi-million dollar oilfield, the room for error is small. We took established first-principle petroleum engineering concepts, combined it with automation and ML to handle large datasets with thousands of wells and different production environments, and wrapped it with a layer of technical NLP so that Nesh can understand oilfield jargon. In keeping with our open architecture, we provide an API that empowers our clients and the other software vendors to integrate their own capabilities into Nesh. Before Nesh, users had to be experts in Petroleum Engineering or Geoscience and conversant with the software UI to get their answer. With Nesh, they can get their answers if they simply know how to speak English.

On a typical day in an Oil and Gas company, during a technical committee meeting, if one of the VPs asks, 'how much oil did the antelope asset produce last month?' it can take up to one day for him/her to get an answer to this simple question because the production data is in some database and the engineer who can get to this database is not in the meeting and the software that can read this time series data can only run on that engineer's laptop. This is a simple use-case but instances like this happen every day and the delay adds up. With Nesh, they can answer questions like this and much more, almost instantaneously.

What are your goals?

As a young startup, we have a few goals. Creating a great product that our users love is at the top of the list. What we have created is a novel product so far and we are looking for strategic partners who can help us in providing guidance and feedback with the next wave of functionalities in Nesh. Our ideal strategic partner is a mid-size to a small E&P company that is willing to try a futuristic piece of technology. Our long-term goal is to create an ecosystem of workflows around Nesh so that everyone in an Oil and Gas company, all the way from an analyst to the CEO, can ask the tough questions and get back the answers without any barriers, delay or learning curve.