Case Antiques, Inc. is an antique and art auction house based in Knoxville, TN. The company has satellite offices in Nashville and Kingsport, TN. The company reaches an international audience with its cataloged auctions. Case Antiques, Inc. specializes in investment quality art, jewelry, furniture and historical items.
In my role as Information Technology Assistant, I was responsible for managing multiple aspects of the company's internet and hardware infrastructure. My primary contributions in this position related to software and process development. However, I also gained valuable computer hardware and network management skills.
My largest project consisted of both an algorithmic and executional restructuring of the bidder approval process. The bidder approval process is often a large undertaking, as it takes multiple employees multiple weeks to query each relevant database and process bidder information. Despite this large investment of time, the company still experienced a high number of unpaid invoices. In order to reduce the amount of non-paying bidders, I designed a new algorithm with which to screen potential buyers. Taking into account multiple factors relating to financial history and personal assets, I was able to use this algorithm to make fast, final decisions on bidder approvals.
|First Name||Last Name||Username||Address||Member Since||Previous Bidder with Case Antiques, Inc.||Value of House||Open Dispute||Dispute Under Last Name||Dispute Under Address||Dispute Under Username|
|Jenny||Smith||jsmithUTK||8888 Example Ave., Knoxville, TN||11-Jun-19||No||57574||FALSE||TRUE||TRUE||TRUE|
|Gabbi||Thompson||gthom098||8765 Sample St., New York, NY||29-Jun-19||No||586271||FALSE||TRUE||TRUE||TRUE|
|Bob||Jones||jbtth7||8765 Park Pl., Atlanta, GA||8-Jan-09||Yes||145594||FALSE||FALSE||FALSE||FALSE|
|Jake||Ruby||jruby||0964 Boardwal Ln., Hollywood, CA||7-Feb-07||No||1297508||FALSE||FALSE||TRUE||FALSE|
Once this algorithm had been developed and its manual implementation tested for efficacy in a live auction, I worked to mechanize it with Python. I developed a Python program to extract the necessary data about each potential bidder from multiple different databases, combining that data into a centralized repository. After the creation of this repository, I automatically executed the algorithm I had previously developed. The program summarizes the results of its findings in an Excel spreadsheet, which is then distributed across the company. This automated bidder approval system proved quite effective in both reducing unpaid invoices and the time required to screen bidders. An entire queue of 400 bidders can be analyzed in under 5 hours, as opposed to 3 weeks. Pre-algorithm non-payments averaged 8-10 per sale. Post-algorithm, the average has dropped to under 2 per sale. As a result, this system is now a standard part of the company's work flow.
Alongside this software development, I also completed projects relating to the company's website design and physical internet infrastructure. In regards to the website, my major contributions were a full redesign and the creation of new marketing media. As far as the network is concerned, I setup an IP-based security system and reconfigured the company's internet hardware.