Ethical Consumer
Application which allows users to search for company executives by NYSE ticker and then click on an executive’s name to see all of their federal political donations since 2007.
•Employed two different APIs to get executive names and their political donations.
•Utilized Redis to cache retrieved information to avoid costly API calls.
Bad Apples
Inherited a legacy codebase and rebuilt the back end to handle production-level scale.
•Utilized NGINX with a custom configuration to load balance between four service instances deployed on AWS EC2 which reduced 1000rps latency from 800ms with a 1% error rate to 14ms with a 0% error rate.
Implemented caching using Redis with lazy-loading, auto-expiration of cached data and LRU eviction policy which reduced data lookup time of cached data by 1300%.
•Exercised data-driven decision making using a custom script to test database query speeds of MongoDB and PostgreSQL.
•Generated 10 million dummy records and seeded them into PostgreSQL to simulate production level data.
•Deployed with Docker to ensure consistent application behaviour between local and deployed services and to easily horizontally scale service.
Jeff's List
Constructed and deployed a product reviews module based on Amazon’s reviews module.
•Used conditional rendering to filter reviews based on user input. Allowed for sub filtering and retained overall filtered order when sub filtering was removed.
•Utilized React-ChartJS-2 to dynamically generate a stacked bar graph representing the distribution of review ratings for each product page based on review data retrieved from custom API.
•Deployed a proxy server on AWS EC2 which combined three other services deployed by on separate AWS EC2 instances.
