This past summer, I interned at ANB systems, a startup that develops innovative workflow management solutions for the energy industry. This was an amazing opportunity to work in a startup environment, as the office was located in a WeWork, where I got to interact with employees and entrepreneurs from various startups in many different industries. Although I had a few smaller projects, I spent the majority of my time designing and implementing an indexing and search service to provide intelligent responses to customer queries and allow new, efficient ways to retrieve and gain insights on workflow data for clients. I was in charge of this from the initial brainstorming, to specifications, implementation, deployment, and testing. I chose this experience to fulfill my 300 hour requirement because I wanted to have the opportunity to experience the intersection of entrepreneurship, innovation, and engineering.
For my project, I was faced with several problems that affected the design and constrained and eliminated some of the traditional solutions involving implementing search. One of the biggest considerations was the nature of the company’s platform was modular and based around configurability and as such, the data was asymmetrical. There was no pre-defined schema that the data could be indexed against, as the data formats were customer defined and dynamic. I worked around this by creating and deploying a service that intelligently creates models of information that work with ElasticSearch, the engine we chose.
Another innovative concept was creating an intelligent search experience that catered to various query forms from clients and worked with context to create the most relevant results possible. In the age of search engines such as Google being so ubiquitous, it was important to us that the search engine be able to match misspelled words, utilize synonyms, and search through various data types such as dates, and numbers from input strings. I worked towards these goals by carefully managing how the service generated the models and indexed data into the backend search service to enable such functionality, while being cognizant of resources and cost.
One specific failure I remember was getting the service I deployed to communicate with another service on our cloud platform. With the help of at least 3 other coworkers, we periodically worked on debugging this issue for almost a week, with no success. I was finally able to track down an obscure security setting that was preventing the network requests from going through. Through this, I learned and solidified my strategy for finding solutions to problems by learning to research tangentially relating topics and following up on clues that sounded similar to the main problem.
With regards to I&E, I think the most important things I learned through this experience was the considerations that are made when making decisions, whether it’s a business or product decision, or implementation detail in engineering. Keeping the client’s perspective in mind, and always being aware of the costs of resources and manpower were some of the things I learned through the various meetings, both formal and informal with both company members and other entrepreneurs at the WeWork we were located at.