My article relating to Peer-to-Peer Lending is one of my most popular articles with over 6,000 visits. I think that's a rather huge accomplishment for a personal website with no desire to publically broadcast. One of the items I talk about later in the article is how to reduce risk while increasing rewards. Using backtracked data that spans billions of dollars, I decided to test out a few algorithms that could potentially maximize my return interest.
After testing these algorithms for over a year, I was astonished at the results. I'll share these in a moment. I thought to myself, "what if I can share this with the world?" It was a unique challenge for me. I knew that Lending Club offered API, as I discussed in the aforementioned article. What I wanted to know was whether or not it was possible to interact with other people's accounts and share the same algorithm using that same API. It was. While challenging at times, I believe my efforts were worth it.
I came across an interesting scenario today while I was building an application that required me opening a file, reading it line by line, and importing it into a dataset. The reason this was an interesting scenario is because there is data that belongs to a specific identifier on different lines.
My immediate reaction was to read the data line by line through Streamreader. However, I discovered that only certain data would get entered. While Streamreader works extremely well for loading an entire file, it doesn't work quite well if you need to read each line individually. It was skipping lines as it read the file.
I came across a database that has suffered a lot of deadlocks recently. This, along with slow query execution among other problems led me to convince management that the database required downtime. Some of the maintenance I performed was re-building tempdb tables and moving them off the C Drive, which I highly recommend for various reasons. You can read more on that here.