It’s been a month since I started my new job! I am working as a Senior Machine Learning Engineer for a small Vancouver start-up named Bio-Conscious. Our main product is Diabits, which is an iPhone app used to monitor and predict blood glucose levels in diabetic patients. My job is to make sure that the predictions are accurate, which is really important since people with diabetes really need to pay attention to their current blood glucose levels and correct those by eating or insulin injections if necessary. Being able to know what the expected levels are within the next hour should make it easier for the patients to plan their actions.

I am really happy to be working on a problem that is meaningful and hopefully can make a positive difference in the world. While looking for work, I considered a number of directions and even had some tentative offers in areas that may be lucrative, but, at least in my opinion, don’t really contribute anything to the society. Examples of such areas are stock market predictions and especially cryptocurrencies. While I do think that blockchain is an interesting mathematical idea, and perhaps currencies of the future will be based on it, I am sure it will not be in its current form. Right now, the way I see it, Bitcoin and other cryptos are essentially a scam and a quite dangerous at that considering the amount of energy necessary for mining them and the associated carbon footprint. So thanks but no thanks, this is not something I want to spend time doing even if I knew that I could make good money with it. I have already spend a lot of time in my life just making money for the sake of it.

I have been pretty busy at work, both because I had to learn some things about the current code base and the way things are set up (for example, I had never worked with Amazon Web Services before), and because the problem of predicting the future values of the blood glucose, like most real-life tasks, is not an easy one. I am not going to go into too much detail, but there are many things to consider, such as the current and past values, the heart rate and physical activity data from wearable devices like Apple Watch and FitBit, as well as other available data. In a way, this is actually similar to predicting the stock market, with so many unknowns and factors that are difficult to quantify. I am hoping to do a good job on it, and I am pretty optimistic about the future of our company.

I am hoping to have time to write more in the future, although I guess we’ll see how it turns out. I also want to keep learning more about the new research in machine learning and attending meetups, which I’ve been neglecting a bit lately since I don’t need to network as much now that I have a job. Anyway, things are going well, and I hope it will continue to be this way.