0 | 2018
As the last digit of this year is almost about to tumble on to the next, I thought I’d take this time to try and connect the dots I’d encountered in my journey over the past year. I’d probably be on the other side by the time I finish this longish journal. This, dear reader, be warned, is more an act of reflection than exhibition. You are, however, welcome to see what the mirror shows. Its gonna be long so you might read a bit, pause, leave and come back later to continue from where you’d left (that is, if you’d still be interested). I have tried to organize them month by month. The theme is Zero. This is what I brought to the table and reduced myself to in the hope of leaving soon with One :)
Some-time, mid last year, I realized that I’d been bitten (again) by the proverbial bug. Some call it the entrepreneurial bug and some others call it the mid-life crisis, I prefer to call it merely a rash creative urge. After having spent almost a decade in helping build Eko into a vibrant fintech platform, impacting millions of customers, I felt pretty spent and began to dream on.
September . I think end of September was when I finally bid farewell to my wonderful colleagues at Eko with nothing more than a faint idea of what I wanted to do. It was hard to leave the comfort zone of a great set of colleagues, a more-or-less stable product, a wonderful set of services impacting thousands of people every minute, and a predictable life.
The most difficult part was preparing my family for the foolishness I was about to embark upon. But to my surprise, they were pretty supportive of my decision. Looking back, they probably hadn’t really realized then how tough this was going to get. Jumping without a parachute? Tick. Family as well? Sure! Tick. Entrusting oneself to a higher power/ destiny ain’t easy. Trust me on that one.
October  is when I hit my first set of zeros. Zero office-home commute (it was a 40 min drive each way over a nice highway and a hellish toll-gate at Gurugram) and Zero salary SMSs :| We had decided that it was best if my better half, Anju and our two kids could spend about a year with my parents in the state of Madhya Pradesh- MP (the heart of India) where they could continue their education without breaking the bank and the kids could spend some quality time with their grandparents as well. So we went on a long drive to drop them there and I spent some time there before returning all alone to my apartment in Gurugram.
All excited about working with zero ‘distractions’ (after a while you begin to miss the distractions!), I set about reading and studying about the field that I had chosen to work on - embedded electronics or its more glamorous counterpart called IoT (Internet of Things).
I was interested in the intersection where hardware met software and magic happened. I was mesmerized by how accessible this field had become for an enthusiast and was encouraged by the loads of learning material available online.
My idea was to revolutionize the way we consumed stuff. I could see many gaps in the entire ecosystem of e-commerce and consumption and how smart devices could fill them and make them seamless and friction free. I spoke to a few people on the use cases and they were excited by it but were not convinced if this could indeed be pulled off seamlessly. So I decided to build crude prototypes, just to test the concept out. Of the many technologies that existed out there, I decided that BLE or Bluetooth Low Energy was the way to go. That is how I learnt about ARM’s mbed platform and used it to successfully commandeer a Nordic Semiconductor nrf51822 module. I also learned 3D modeling and discovered 3d hubs to get my enclosure printed. To give you a sense of my satisfaction, I need to tell you that I had to start from zero. It had been over a decade since I had done any real programming and even more since I had touched electronics!
November . What hit me hard by now was having zero family around. I really missed the hugs, the fights, the sounds and the noises that I used to take for granted.
By mid November, I had a kinda working prototype! Anyways, with the prototype, I contacted a patent attorney and found that the concept already existed in the form that I had made it. At this point I felt even lower than zero! So, it was back to the drawing board for me :(
Another thing that happened was that I had started to cook (not that I had much choice). No fancy stuff, just simple food that could be put up quickly. Here are a few samples from my November folder :)
December , was when I took some time off ‘work’ to record a single that two of my friends and I had written and composed some 15 years ago! The song was a prayer called ‘Keep My Faith’ and it is the theme for my journey :)
I also took some time off to visit my folks back in MP. Then spent a week in Kerala for our high school reunion and was back again in Gurugram for the new year.
January . I spent most of my time thinking about the product variant that could achieve similar results to what I had prototyped earlier but still be novel and unique. After much thought and intense day-night experimentation marathons, I put another prototype together and contacted my IP folks. They said it looked promising and we got about verifying and filing a preliminary patent in due course.
Edit: Hat-tip to the team at Khurana&Khurana :)
February  had me visit Kerala for a family occasion from where I traveled to Bengaluru to meet up an acquaintance from the VC world who might be interested in guiding me with what I was trying to put together. That demo went OK, but I guess things were too premature and unprepared for him. I also planned to visit a couple of my friends who were also in Bengaluru then. One of them was my friend and classmate from REC/ NIT, Abhijeet Mankame.
After having shown him my crude prototype and attempting to ‘sell’ my story, he said something along the lines that it was probably expecting too much too soon, he said that he knew of some other pressing problems in the FMCG (Fast Moving Consumer Goods) world that could do with some IoT and innovation. He explained how real time insights were missing from the last mile retail in India especially due to the fact that the producer company was a few layers disconnected to the final retail shopkeeper. While this abstraction brought operational efficiency, it meant that most of them only had post-sales data to rely on, to clearly analyze the impact of a new product or campaign. Could we make smart devices for these mom-n-pop retail outlets that would give this insight? If yes, he could put me in touch with a few key people from one one of the largest retail FMCG companies in India. Sure, said I and put together a smart display unit for a demo meeting within a couple of days.
It would play the product jingle every time a customer lifted a product from it and would also update an online portal in real-time on the number of SKUs left on it. This would give the company a birds eye view of how the product was doing in different locations and even give insights like which time of the day were maximum sales happening. Abhijeet arranged for a 30 min slot with this busy senior executive at this FMCG co. The demo went well. They were interested in the concept and that got me a PoC at one retail outlet for one of their products. Lil victories, thanks to a friend :)
Nitin Garg, another friend meanwhile was trying to set up a niche digital lending platform. After a few discussions, I decided to help them put together an architecture. I guess I worked on this in parallel, just so that I could also continue to keep in touch with the fin-tech world that I had just left and also to ensure that I could make use of the decade long experience I had in the sector.
I think I did deliver them the first version in about a month or so. I continue to advise his company Flow Global on technology.
March  took me back to Gurugram and we had a small family reunion before they finally left for MP again after a few days :(.
Almost a month and a local metal shop later, I had a better prototype ready for the PoC.
By the end of March I got back to Bengaluru and had this deployed at one of the Village Stores here. This was a WiFi based device and was made on the ESP8266 module programmed to very low current consumption that should have the unit last for about 3 months on a single charge.
By now, I had some other prospective clients showing interest for similar retail IoT product and thought that I should probably formalize the existence of my startup. Anju and I had a short-list of names SmartThings, TronPowered and SmartBees. We bought a few of those domains.
I like bees. They are hardworking, intelligent, tiny but powerful beings. They work tirelessly in the background to sustain our ecosystem. They reward themselves with a little bit of honey (rhymes with money :) for each endeavor. And they carry with them a nasty sting for those who try to mess with ‘em. Pretty much the character I had in mind for our digital avatars. And no, we hadn’t seen the Black Mirror back then. So SmartBees it was, for a while.
Anju, by then was part of a unique lab called the Atal Tinkering Lab. Read more about that here:
The word tinker was nice. It sounded more positive than a hacker and had a nice sharp musical ‘ting’ to it. One fine morning, I knew I had it. TinkerBee! I called up Anju and she liked it too. Thus, TinkerBee was born.
April  is when we attempted a hand at branding. Some early sketches:
The consensus among the people I consulted and my family lead to the following:
It had t for tinker and b for bee. Looked like a bee. Had a point source radiating wireless. Also looked like a guy with a beard and a mohawk. Gave this to my cousin Daniel James who is a creative professional with an agency here and he gave us a few really great options, and a refined version of the one above, which is what we use now. So thanks to Dan for our logo :)
April is also when Anju and I got our digital signature tokens as is mandatory for directors in a company in India. We used SignKaro.com and it was a smooth process. I’d recommend them.
It also marked the beginning of a longish and costly journey trying to get our company Tinkerbee Innovations registered as a Private Limited company in India. It took us over a month, especially since we did not have a lease agreement/ property in Bengaluru. All in all, we would have spent around Rs. 50,000 just for these formalities and paper-work.
My friends Sanjana and Jayadev Gopalakrishnan introduced me to IKP Eden. It is a wonderful incubator/ workshop/ co-working space located at Koramangala, Bengaluru, focusing mostly on hardware, software and biomedical startups. They have a nice workshop on the ground floor with metal-work, carpentry and 3d printing tools. Helps get quick prototypes of the hard kind out. And they have a shared space for work in the second floor. They also have a good biomedical lab on the third floor. I’ve been in this place since then and it has been good so far. Not flashy like WeWorks and the ilk but nice, functional and efficient.
This was also when I had to learn to move from a prototype on a general purpose PCB to a more professional and reliable PCB design. Thats when I discovered EasyEDA and JLCPCB and I’ve been using them ever since. Started from basic circuit designs and can now do super compact boards with, high quality SMD components and RF optimized designs.
EasyEDA - Online PCB design & circuit simulator
EasyEDA is a free and easy to use circuit design, circuit simulator and pcb design that runs in your web browser.
Helped get the designs from:
With the help of a few friends, we also managed to put up a dashboard to gather and visualize the data that these “StockBees” were transmitting.
It was a basic php site hosted on goDaddy and used FusionCharts trial version.
May 8th 2018 is when Tinkerbee Innovations Pvt. Ltd. got formally incorporated. My family and my parents too had come down to Bengaluru and we went on a trip to Goa and North Karnataka, visiting places like Honavar, Karwar and Ankola. And we were back in Bengaluru to attend the wedding of one of my cousins. And also visited Kerala.
Not much work got done. Except that a set of 4 StockBees were now in a store in Bengaluru, silently monitoring the stock movement of one of the test SKUs.
June  was spent on more learning and R&D on reducing the cost and improving the efficiency of the original product that we had started out with. So that entailed better circuit design and improvements in the firmware for the BLE code.
A lot of time was also spent on numerous meetings with shop-keepers and the people who worked there, in trying to understand and document their concerns, needs and context while trying to design solutions that addressed the data gaps that existed there. India is a unique country where over 95% of the retail outlets are in the ‘unorganized’/ general trade segment with the more organized modern trade segment fast catching up. One thing that was obvious was that things that worked elsewhere may not work here as is. The solution had to be tailored for this context and had to address the UX concerns of all the stakeholders involved. This wasn’t as easy as I had earlier envisaged it to be!
July  was when we tested out our really compact EyeBee unit and had managed to reduce the cost and size significantly.
It is also a time to remind you, dear reader, if you have reached thus far, is how faithful I’ve been so far to the theme zero. The journey so far, had a lot of learning and costs but zero in sustained revenues :). The original product that we had started with still has zero traction. But there was a sense of a product-market fit for StockBee, the accidental product!
July was also the time I discovered LoRaWAN (I will write about this technology in detail in a different post, hopefully soon). While WiFi was working fine, for our initial pilot, it had a few drawbacks. One, it needed a source of mains power. Despite our best optimization, we couldn’t get an 18650 lithium ion cell to power it for more than 3 months at a stretch. Two, it needed a proximate wifi access-point in each store. This was a costly issue, both in terms of capex and opex at scale and required the procurement of SIM based WiFi dongles.
LoRaWAN enabled compact end-nodes that could be battery powered, could trasmit for over 3 Km in urban conditions and would last for 5 years atleast and 10 years at max.
Most importantly, this provided a zero UX solution for the shop-keeper. I just had to place a passive unit in the shop and it would need no power or intervention from the store owner. We had successfully reduced the ‘solution’ to a mere point-of-sale prop, almost like a passive mat for them to keep the SKUs (Stock Keeping Units/ an individual product being sold) on.
August  is when after much experimentation and tweaking, we had our first LoRaWAN gateway and node up and running.
Mid August is when I had to travel to Kerala to attend my cousin’s wedding. Unfortunately, that was when the great floods struck and I was caught, with my in-laws, in it. Those were a few really tense days. Another grim reminder how in a flash, nature and powers above us could reduce us to zero. All that we’ve built, enjoyed and dreamt of, could simply be washed away.
It was also an opportunity to realize the tremendous power and the inherent fallibility of technology. For days, there was no electricity, no celllphone connectivity, no digital money, no ATMs even to spew out physical money, no transport and limited supply of food and water.
It was good to see humanity coming around to support those who were in need and I am indeed thankful to all those who directly and indirectly contributed to the relief work. The only sad take-away is that systemically, we are still unprepared to face the next disaster. I’d rather that we prepare, like Japan does for earthquakes and typhoons, than scramble around later.
August was also when Rony reached out with a control and configuration software requirement for his company Precise Displays. The project continued for over a month and we were able to put together a nice Windows based platform for them. Thankful to him for the work!
September . Based on a meeting with the FMCG co for whom we had deployed the PoC, we had now got an intent to expand to a slightly larger paid pilot in around 60 retail outlets. So September had us scrambling to design the smart trays which could be sturdy and could be scaled. Also, we had moved our board from WiFi to LoRaWAN due to its obvious benefits. The following is the result of those endeavors.
I must also highlight a few other significant zeroes in my story. From May onwards, we had applied to about 12 accelerator/ incubator programs and have met 10 key people in the venture investment space. We were rejected/ politely declined by all of them :)
No, I haven’t yet exhausted the list or given up hope as yet, but just had to let you know that I could now tick the ‘rejected by a significant few’ box in my startup journey checklist.
Abhishek Sinha, co-founder and CEO of Eko and Piyush Gupta, our friend and colleague there, who is now in the valley, have been a constant source of advice, support, direction and feedback. Another person who has helped me a lot is Raunak Singhvi who was also a colleague at Eko and who is now with Grey Matters Capital. I am indebted to them, but I still continue to be foolish despite their good advice :).
October . The Indian government has instituted a few nice support programs for startups and one of them is the Nidhi-Prayas grant. IKP Eden, which is an incubator supported by the government was a participant in the grant program. So October saw me pitching to a group of eminent people brought together by IKP for a grant towards taking our StockBees from prototypes to a scale-able product. Thankfully, we were selected and my current journey, atleast the design and product development part has been powered by them.
Most of the rest of October was spent on R&D on LoRaWAN, with about 6 main revisions and twice as many sub-revisions to our original hardware board.
Edit: This was also the time that I got in touch with eLagaan. Hat-tip to them as well. They’ve been handling all the CA and accounting tasks for us!
November  was spent trying to finalize the set of component vendors, comparing product BoM costs, finding out electronic assembly partners, sourcing from China, antenna suppliers et al. I was introduced to a designer Saif Faisal for the hardware design by my friends Rony and Lisha Stephen.
Another old friend who has been continuously helping me fine tune my solution and direct me to the middle east market has been Manish Bhandari. I had the privilege of working with Manish briefly at Eko. I learnt the art of follow-ups from him, but am still at the intermediate level :). Immensely thankful to him for all his help!
December . More R&D on LoRaWAN and more testing. Its been a challenge getting the antenna right.
Have been testing a few different options available out there. We seem to finally have got the pieces together for scaling up the product. Also, signed up with Tata Communications as a startup partner. Had earlier tied up with SenRa, but Tata seems to have a much wider coverage for atleast my initial pilot. But their tech platform is way behind what even a free platform like TheThingsNetwork offers! Hope to get this part sorted out soon, one way or the other.
Just waiting now for the first production cycle runs and then testing. And that brings us to the end of the year and time has finally caught up with this looong post as we are headed towards the first evening in January 2019.
When I took the plunge, almost a year ago, I had not realized that there were so many friends and family who were ready to go out of their way in trying to help out. Jayadev, my brother Arpan, Arpitha his better half, my aunts and uncles in Bengaluru, my parents and my in-laws have been gracious in providing me a shelter and support when I needed. I would like to thank them all for having stood by when I was at my zero. I thank God for them. The people who have suffered the most are my wife and kids. Hope we can make it up together!
Here’s to wishing you all a great year ahead. Lets hope and pray that we will inch past our zeros as we step into a new year.