Pod Insights The Brand Called You Revolutionizing Battery Diagnostics: Shubham Mishra, Founder & CEO, Energy AI Labs About this episode Welcome to another episode of The Brand Called You featuring Shubham Mishra, Founder & CEO, Energy AI Labs.
In this enlightening conversation with Ashutosh Garg, Shubham shares his entrepreneurial journey—from building E-bikes to pioneering cutting-edge battery diagnostics in India. Discover the challenges of Deep Tech innovation, the evolution of EV Doctor, and the vision behind creating a universal Energy OS for homes and businesses.
Shubham also delves into battery myths, the impact of AI and field data on diagnostics, and what the future holds for Energy Super Intelligence.
Whether you are passionate about EVs, AI, or the sustainable energy revolution, this episode delivers practical wisdom, fascinating stories, and a glimpse into the future of how the world will produce, consume, and value energy.
🔔 Don’t forget to subscribe and hit the bell icon for more inspiring interviews with leaders and innovators.
Transcript 0:08 Speaker 00 Welcome to another episode of The Brand Called You, a podcast and podcast show that brings you leadership lessons, knowledge and experience and wisdom from thousands of successful individuals from around the world. I'm your host Ashutosh Garg and today I'm delighted to welcome a successful entrepreneur with a deep understanding of AI, batteries and EVs from Ahmedabad, India, Mr. Subham Mishra. Subham, welcome to the show.
0:34 Speaker 01 Hello sir, thanks for the invitation.
0:36 Speaker 00 Thank you. Subham is the founder and chief executive officer of Energy AI Labs. So Subham, before we start talking about Energy AI Labs, which is going to be the focus of our conversation, a few questions for you. You began with a simple problem, battery diagnostics. What was the moment of frustration or insight that made you say this must be solved differently?
1:01 Speaker 01 Okay, so we started off with e-bikes actually. So right after my job at GenSol, okay, I was at this part of the design engineering team that I left it within one year and started this venture named the company E-Vega Mobility Labs. Vega is the Sanskrit word for velocity. So that was the initial inception part. We built e-bikes and we were constantly building them for various purposes. We were clueless how to sell them, although we are, we were making them. So in 2020, we saw one opportunity that why don't we keep suggesting EV conferences, events and exhibitions?
1:44 Speaker 01 And we will talk about what we have built. If there are any prospects, they will be getting converted and we will get some sales out of it. But COVID hit, okay, and we were clueless again. Now what to do? Now no events or exhibitions will happen. So we directly reached out to Nobel laureates across the world, okay, related to lithium batteries. And one of them replied suggesting some resources, some research papers and conference papers to read about. In that journey only, we met one guy from Gandhiragir, that is the capital of Gujarat. We met one guy, he was EV mechanic.
2:24 Speaker 01 He has, he had one technician who was putting his bare hands on battery pack surface. Okay, so when the battery got hot, uh, he directly inferred that, that this battery is bad. Okay, so that was the moment of frustration or, uh, you can say a eureka moment in hindsight for us. Where we have put, uh, where we have seen that putting bare hands on battery pack surface is the norm for getting the battery test done. Okay, so just like humans check fever, they were checking.
2:55 Speaker 00 Yeah, amazing. And how did working with Ajay Vaswisth with a defense R and D background, shape your technical philosophy and execution discipline?
3:06 Speaker 01 Uh, yeah, so I'm quite a chaotic guy. So initially I was, I didn't know about the systems and processes. How structure will help. So he brought the structure because he has worked in the systems part. He has worked as a part of a, a deep, uh, R and D based hardware based system. He corrected the path on various fronts, uh, right from the hardware, T R A levels to how to shape the customer journey. In parallel with that, I was making sure that, uh, we are not, uh, too much structured during that process because at times over structuring directly like, uh,
3:48 Speaker 01 constraints the chaotic part and it limits us from thinking outside the box. So, uh, he brought the structure, uh, I was part of the customer, uh, understanding and empathy field. And together we have made sure that, uh, the rational and rational, uh, they perform better in an M L V material.
4:09 Speaker 00 Hmm, very interesting. And, uh, one more question before we come to Energy A I Labs. Deep tech entrepreneurship in India is not easy. What were some of the hardest non-technical challenges such as funding, talent, credibility that nearly derailed you?
4:29 Speaker 01 Yeah, so funding, yeah, it was a problem, but later on we found that for our product, it is different for different products.
4:37 Speaker 00 Yeah.
4:37 Speaker 01 But for our product, millions of funds were not required.
4:42 Speaker 00 Hmm.
4:42 Speaker 01 Uh, we have to make sure that the algorithms are right, the testing accuracy is right. A lot of times, a lot of money is wasted just for getting that accuracy right and that was not favoring our mindset. Talent, yes, uh, being from Ahmedabad, we were not able to get top A I M L engineers uh, who have worked on product, who have worked on deep tech, so that was one part of it. And last one is the credibility part, that was the biggest one.
5:11 Speaker 00 Hmm.
5:11 Speaker 01 Uh, Indian, Indian people trust imported products more than uh, Indian product system. So that was the biggest uh, global, you can say, where we have to keep uh, working on the learning curve.
5:26 Speaker 00 Hmm.
5:26 Speaker 01 We have to constantly prove our credibility, trust and the merit that we have built through this product. So that was the highest, like largest part, but uh, on uh, the opposite side of it, when the trust is built, it has a flywheeling effect. And it directly uh, uh, got us a lot of success and credibility.
5:48 Speaker 00 Hmm, amazing. Uh, Shubham, let's now talk about energy A I labs. You were just telling me a few minutes back about how uh, you know, someone would touch a battery to determine uh, the health of the battery. Uh, before E V doctor, you know, it was largely guesswork. But tell me, why did the ecosystem tolerate such low scientific rigor for so long?
6:16 Speaker 01 Uh, yeah, so this is very interesting question and even we got the answer after working on it.
6:22 Speaker 00 Hmm.
6:22 Speaker 01 After serving it to like a lot of people. So in India, diagnostics is only happening when something is wrong.
6:30 Speaker 01 Okay, people hate diagnostics.
6:32 Speaker 00 Yeah.
6:32 Speaker 01 Any type of diagnostics when some issue is there even with our own health. If there is an issue, then we will realize the importance of yearly checkups and everything. Right, so this was going on, bulky machines were imported, they felt like you have a machine.
6:48 Speaker 00 Hmm.
6:49 Speaker 01 And they were complex, so they were creating a lot of FOMO also. That this is the machine that checks the V battery and you have to sit there for eight to ten hours. You have to get the test and once the test is done, you will get the report and sometimes if you miss any readings. Uh, you will be directly having non-reliable report, then you have to check it again another seven to eight hours.
7:11 Speaker 00 Hmm.
7:12 Speaker 01 Uh, diagnostics not being, not being the primary concern was the reason. Uh, eventually what we have found that if it is convenient enough.
7:22 Speaker 00 Hmm.
7:22 Speaker 01 Then people will check it easily, so just like thermometers. Okay, so if they are convenient and accessible, then uh, whenever somebody is feeling feverish. They can directly use some thermometer.
7:36 Speaker 00 Hmm, very interesting. And yet, you know, the eight to nine hours that you were speaking, now you introduce the fifteen minute diagnostic.
7:44 Speaker 01 Correct.
7:44 Speaker 00 Which is very compelling. What were some of the hardest engineering constraints to compress testing into the timeframe without losing accuracy?
7:54 Speaker 01 Yeah, so biggest problem was the availability of data sets.
7:57 Speaker 00 Right.
7:57 Speaker 01 Uh, there were no public data sets of real-time battery uh, issues, okay. Uh, so uh, first one is the objective part or and second one is the subjective part. So, objective part will cover the data sets, real rational data sets in the structured form and. The subjective part is uh, real insights that we get about the batteries, how they perform. Uh, how different type of batteries perform in different ambient conditions and all. So, we were not having idea of both of them, so we built our own data sets initially. We conducted thermal runaway experiments along with HiIT Gandhi Nagar.
8:36 Speaker 01 Sixty, seventy batteries we have put on fire and at the same time we were getting the electrical equivalent readings. So that we can make sure that the accuracy is right when the extremes are there, okay. So the biggest problem was fire. So we started with fire and eventually we reached out to the point where we are providing fifteen minutes test reports about battery performance, safety faults and whatever goes inside the battery handling and maintenance.
9:07 Speaker 00 Amazing. From your over fifty lakh plus tests, what is one counterintuitive insight about battery behavior that most people get wrong?
9:20 Speaker 01 So, most people get wrong is that batteries will last only, lithium batteries, EV battery packs will last only for three to five years. That is the most counterintuitive thing because many brands have positioned it in such way because they don't use branded cells or they don't use right topology or architecture. They can easily last till eight to ten years if, well, if there are no partial charging behavior, if the components are right.
9:53 Speaker 00 Subham, calibration is critical in diagnostics. What are the core principles you follow to ensure accuracy and when it comes to very, very large numbers?
10:05 Speaker 01 Yeah. So initially we started with benchmarking. Okay. So we took the best of the best tools which were available right from Japan to USA to Europe in China. And right now we are supplying to all four countries. Okay. So that part from benchmarking to being a competitor to them was like a roller coaster journey. What we have made sure is we constantly tested real time batteries. That is the best form of intelligence that we can see. Okay. The intelligence which can act, which can empower decisions, right? We are making sure that whenever any test happens with the EV doctor through any of our clientele, it is accurate
10:50 Speaker 01 enough. So right now it is constantly getting trained on real time data and that's where the real calibration happens.
10:57 Speaker 00 Okay. And can EV doctor eventually move from diagnostics to predictive failure in, to prevent, you know, problems in real time?
11:08 Speaker 01 Uh yeah, so right now it has the intelligence for that, but we don't have the control for that. Okay, we don't have any proper tools or SOPs that we can integrate with our intelligence. So that they can directly follow it as part of the system.
11:25 Speaker 00 Hmm. And for fleet operators, how does accurate battery uh translate into a direct financial impact? Give me an example also if you can.
11:39 Speaker 01 Yeah, so uptime is the answer. So if there are hundred EVs within your fleet. And twenty are dead because of some issues. And if you are checking them and you have found that all the issues were not big enough, only four or five had big issues. Fourteen fifteen of them were good, then you can make them alive. Uh they'll be directly generating revenue for you instead of being a dead stock, they'll be an asset. This is the simple explanation of.
12:10 Speaker 00 Hmm, and uh when you talk of batteries for fleet operators. Is it a very significant chunk of their entire uh the the maintenance revenue or maintenance costs?
12:21 Speaker 01 Uh yeah, uh initially no, but eventually when the fleet explodes. Uh they go from city to city, then it becomes a misat.
12:30 Speaker 00 Hmm. Uh my next question, Shubham, is that insurers and financiers rely heavily on risk models.
12:39 Speaker 01 Correct.
12:39 Speaker 00 How is your data reshaping risk pricing for EVs?
12:44 Speaker 01 Yeah, so we have already partnered with some big insurance players. They have got EV doctor devices at their touchpoints. Uh although it is not commercialized at scale, uh we are like in the nation stage for that. But how this helps them is to underwrite the batteries in an independent manner. Okay, so till date they have been inde- uh they have been dependent on the OEM's data only.
13:09 Speaker 00 Hmm.
13:10 Speaker 01 Okay, and at times it was biased, at times it was uh not correct because of the warranty losses. The abusive handling or some maintenance or repair issue.
13:22 Speaker 01 Through our device, they are sure that a third-party tool is there for getting the battery diagnostics. Then you will get the report and as per that you can do the premium correction. You can do the claim optimization, you can make sure that the battery needs to be replaced or repaired. So, all these decisions can be made through this device.
13:43 Speaker 00 Hmm.
13:46 Speaker 00 Your system blends hardware, AI and field data.
13:51 Speaker 00 Where does your real competitive advantage lie?
13:55 Speaker 01 Uh, right now it is with the field data because hardware, obviously anybody can reverse engineer. Uh, we don't have a monopoly on that and we don't want to have. The main thing is intelligence where we have the field data and we are constantly training our algorithms to become best, to become accurate. Along with that, we are partnering with some key OEMs, uh, insurance players and finance players. Where we get a lot of access to that data where we can perform our intelligence too. So, our intelligence, uh, like which is powering EV doctor is called battery OS. Uh, that is, uh, uh, uh, part of energy I Laps, uh, it's battery operating system.
14:38 Speaker 01 So, till date, there is no unified software for batteries. Our phone is having different, our laptop is having different, our, our timer is having different. So, every battery is having their own software layer. If we can have one, then they can coordinate with each other. They can make sure that you live a comfortable life while not worrying about the batteries.
15:00 Speaker 00 Hmm. You just were speaking about the battery OS. How close are we to building an energy OS? What does this look like in practical terms over the next few years?
15:12 Speaker 01 Yeah, so everything, every big intelligence thing or platform, they have their own operating system as of now. So right from the Android to sound to display, everything has a benchmark or a signature you can say. Right, powered by Herman, powered by Intel, whatever is there.
15:32 Speaker 00 Yeah.
15:32 Speaker 01 So we have found that batteries don't have such uh operating system. Okay, so we are already building one and diagnostics is just small part of it, monitoring alerts, maintenance, uh scheduling, charging patterns, uh all these things, they are part of battery OS and we are constantly building them. We have built APIs for most of them. We are currently working with some OEMs also on pilot levels. Where we are making sure that battery OS becomes the norm.
16:01 Speaker 00 Hmm.
16:01 Speaker 01 Now coming to the energy OS part. So we have seen Jarvis in Iron Man. We have seen a lot of home automation projects. We have seen a lot of smart meters and uh smart plugs. Okay, but most of them have never made it so successful that uh they become part of the default option.
16:22 Speaker 00 Hmm.
16:22 Speaker 01 Okay, they are not as default as switches. Okay, so with energy OS, what we are making sure is to become an obvious option.
16:30 Speaker 00 Hmm.
16:30 Speaker 01 For energy software where efficiency is not just a primary goal, okay. The primary goal is intelligence where devices can talk to each other, devices can coordinate in such a way that your bills are minimized and your comfort is maximized.
16:47 Speaker 00 Hmm.
16:47 Speaker 01 Uh they should not hamper your overall experience. And for enterprises and utilities and industries, it means a lot of E S G compliance uh norms as well.
17:00 Speaker 00 Hmm.
17:00 Speaker 01 Okay, so following E S C compliance norms and directly generating credits while saving electricity in a comfortable way is the way to go. So that's why we are building energy OS.
17:12 Speaker 00 Very, very interesting. How close are we to a world where battery safety is continuously monitored rather than periodically checked?
17:23 Speaker 01 Yeah, so we should be close but uh uh cost is the concern because uh constantly mesh getting the battery safety on your uh display or on your dashboard is uh a costly affair as of now. As we democratize this battery OS, it will be part of the normal thing and it will not cost uh more. Right now, through periodic testing, we are getting a lot of data that will directly lead to APIs and once these APIs become default for any BMS, then battery safety can be accessed just like SOC percentage that we see.
17:59 Speaker 00 Hmm amazing. Shaboom, I have time for two more questions for you. How will energy AI labs move beyond EVs into homes, buildings and decentralized energy systems?
18:13 Speaker 01 Yeah, so uh the inflection point will be inverter systems because that's where the homes and batteries, they both coincide and uh we are already working on it. Uh once we have solved some real time issues with the inverters, we have the plan to build one uh central uh democratized uh operating system where we will cutting from homes, uh not from India, uh from the western world because uh their smart meter data is directly available. Uh whereas in India, it is difficult to get the access of it because in India still the data is fed from the smart meter just for the bill calculation and some uh
18:57 Speaker 01 usage uh data, right? It's the show on the dashboard or on the bills. But more intelligence if it is there, then we can do the demand response scheduling, we can make sure that the time of usage is right, we can make sure the appliance health is good, we can also uh suggest them best appliances for their usage instead of them directly random guessing it or uh purchasing it based on the offers they get online or offline.
19:26 Speaker 00 Hmm amazing. And my last question for you, you know, when I was reading about you and preparing, I have read this phrase, energy super intelligence.
19:37 Speaker 00 What is energy super intelligence and if you succeed in building it? What fundamental shift will the world experience in how energy is produced, consumed and valued?
19:50 Speaker 01 Yeah, so globally, three to four trillion dollars are lost every year just from the electricity power loss. Okay, right from the transmission to distribution to generation. So all these. And in India, it is beyond fifteen percent as of now, although it is reducing year over year, but it is still a lot. So what we are making sure is starting from homes to utilities, to industries, to enterprises. If we can coordinate with each other and if we have made sure that energy OS or any platform which is supported by energy OS intelligence.
20:30 Speaker 01 Then we can build a Palantir for energy where systems can talk to each other. So that the grid is not overloaded, so that the energy bills are in control. So that people can leave their energy related decisions through the, to this intelligence platform.
20:51 Speaker 00 Hmm.
20:52 Speaker 01 So this is the major vision behind it. Uh, in the most layman term, it is like Jarvis on steroids.
21:01 Speaker 00 Hmm.
21:01 Speaker 01 And in the most layman terms for a consumer, it will be like a universal basic thing for energy.
21:10 Speaker 00 Hmm. Fascinating. And on that note, Shubham, I just want to say, what an amazing conversation this has been. I mean, I didn't understand too much about uh, batteries and you know, intelligence needed to be able to understand our own power consumption. But clearly you're doing an incredible job and not just supporting automobiles on how to manage their batteries. But your vision is, is absolutely fascinating on how you think homes, offices, buildings can really manage their energy better. Thank you for speaking to me and all the very best to you.
21:47 Speaker 01 Thanks a lot, sir. Thanks a lot for the invitation and it was a pleasure meeting you.
21:50 Speaker 00 Thank you.
21:53 Speaker 00 Thank you for listening to the brand called You videocast and podcast. A platform that brings you knowledge, experience and wisdom of hundreds of successful individuals from around the world. Do visit our website www.tbcy.in to watch and listen to the stories of many more individuals. You can also follow us on YouTube, Facebook, Instagram and Twitter. Just search for the brand called You.
Previous Designed to Win: Leadership Lessons from S Devarajan, Former Managing Director of Tata Elxsi Next Unlocking Human Behavior in Branding: Paul Larche, Author & Behavioral Branding Strategist Page updated November 14, 2026 · transcript and summary are machine-generated