I created a RESTful CRUD application using Angular's $resource and integrated it with a Java backend using Google Guice for dependency injection. I used JBoss WildFly as the server and employed various libraries like Jersey for the REST client and Apache Commons Lang for utility functions. The project is available on GitHub and deployed on OpenShift, showcasing CRUD operations through simple CURL commands.

I realized my commit messages were weak and didn't convey the intent of the changes. Writing meaningful, committed messages is crucial for clear communication. I researched best practices and proposed standards for better commit messages. These include using a short, imperative mood summary, detailed paragraphs, and logical grouping of files. Good commit messages to speed up the review process, help future maintainers, and aid in writing release notes.

Today, I interviewed a senior developer candidate at OnShift. I value passion and continuous learning over experience alone. Key questions included understanding SOLID principles, designing a small commerce site, technology stack choices, integration and deployment strategies, secure handling of customer data, and proposing new technologies. I seek candidates with a strong grasp of these concepts and a drive to improve.

In January 2015, I shared an article by Martin Fowler on feature toggles with my colleagues, Dean and Bryce. The article explores the complexity and use of toggles for cohort, A/B testing, and feature releases. At OnShift, I recently developed a "feature permission" REST endpoint for beta testing new features in our mobile apps, which I may rename to "feature toggle." We're also using Google Analytics to track user flow and feature adoption.

At OnShift, a Cleveland-based startup, I transitioned from 9 years of Java development to embracing Python. After a month of immersion, I appreciate Python's readability and extensive standard library. I explored the ecosystem through podcasts and compared Java and Python tools, collections, and frameworks. This shift has broadened my development perspective and highlighted the vibrant Python community.

In February 2015, I joined Explorys, a company using Hadoop to manage and sell vast amounts of hospital data. Hadoop's distributed processing and fault tolerance make it ideal for big data despite the single point of failure at the name node. Key points from "Hadoop - The Definitive Guide" include its scalability, cost-effectiveness, and powerful data processing capabilities. The book provides insights into Hadoop's MapReduce framework, HDFS, and ecosystem, highlighting its role in efficient big data management.