Smart contracts and the block chain are great. They offer a transparent way to operate without any human intervention. They come with a downside however. The hash of the block is not known until it is mined. This is a downside because it requires a large amount of GPU power and electricity to solve these hashes. It begs the question, isn’t there a better way to harness these resources? Luckily a new ICO called TensorBit has emerged with a proof of work concept which promises to take these resources and apply them to a good use. That use is solving machine learning and artificial intelligence computations. This is a big claim, and one that deserves a true analysis to see what this is all about.
In the past analyzing ICOs, one of the largest factors to the success of the company is who is behind that company. I like to look at these ICOs the same way that Paul Graham of Ycombinator (the American seed accelerator that was behind companies like Airbnb, Reddit, DropBox, and Stripe amongst others.) might look at the companies he is going to invest in. With that in mind this analysis is going to focus on the founders of TensorBit and Paul Graham’s guidelines for selecting good founders. (http://www.paulgraham.com/founders.html)
TensorBit is entering a new business world using cutting edge technology like the blockchain and machine learning. The founders better be determined to make this a success, otherwise I can see this being quickly overwhelmed with competitors ready to eat their lunch. While the http://www.tensorbit.net website does not completely cover the founder’s backgrounds, they do have some interesting Linkedin Profiles. It looks like the founders are highly educated and already working within the STEM industry. I see a good mix of math, computer engineering and data science. This certainly proves they are smart, but I am not sure it answers if they are dedicated. I thought to get the best feel for this company, I would need to talk to them directly. I was able to get a dialog with one of the founders, Luuk Jansen. Luuk is the data scientist in the group and told me that he was the one who had the initial idea behind TensorBit. After going back and forth on email I started to get the impression that Luuk was far more dedicated to TensorBit than I initially imagines. Apparently, he has been trying to come up with a concept to speed up machine learning using distributed computer power for over 4 years. Luuk went into details behind some of the obstacles in creating this and I got the sense that this guy was serious, and he had the drive to really make this a success. After speaking to Luuk, I can confidently say that TensotBit certainly meets this requirement.
I mentioned earlier that TensorBit had a solid ensemble of founders. Covering all the areas in STEM. I believe this lends to the flexibility. They can easily pivot their business idea based on the skillsets of the team. There should not be an engineering problems they cannot solve, any data problems, or anything mathematic. This is a very good and further promotes that TensorBit will be successful.
This is going to be an easy one to evaluate. I will go out on a limb and say that any individual that can look at different pieces of technology and combine them in a way that has never been done before takes a lot of imagination. That is exactly what the founders here have done.
This is kind of a strange category to check against, never the less the successful Paul Graham swears by it in predicting the success of a startup. I am going to go back to the conversation I have with Luuk, he explained to me how he encountered some obstacles in the beginning of refining his idea. When I asked how he got around an obstacle, which was the difficulty in designing his idea into a proof of work. Luuk explained how he had hacked the block chain to gain some advantages. When I say hacked I mean coming up with creative coding solution to get to the ultimate objective, not hacking into computers. In Paul Graham’s eyes, this shows naughtiness, so it does for me as well.
I must say I would not have been able to answer most of the questions without having the opportunity of speaking to Luuk. I was told that all the founders have backgrounds together and have worked with each other at previous employers and have always clicked. This is a great sign as that means we should not be seeing any disgruntled arguments about the way the business should be ran. These are not fly by night founders, they have the history, experience and knowledge to all work together nicely.
After reading the TensorBit whitepaper, speaking to Luuk, and running my typical analysis on this hot new ICO, I am thoroughly impressed. TensorBit is about to change the blockchain world with the proof of work to start utilizing all those miners and GPUs to start solving real problems. Problems that also happen to be very valuable for large tech companies. If TensorBit can deliver on its promise to offer a machine learning and artificial intelligence service which cuts down on the time required to train models, they will change everything. This is one worth looking at, the technology is amazing and the founders are meeting the same requirements someone like Paul Graham would consider.