Python developer

General Software Development Seattle, Washington, United States

Profile
Contacts

Summary

While working for Amazon and TCS, I prided myself on making sure that I always kept the customers' needs and cares at the top of my mind. For instance, I created a TTL (time to live) automation system for registering vendors on the Amazon marketplace, which saved about 15% of storage space for our vendor customers. Furthermore, I am willing to go above and beyond to ensure customer satisfaction, including providing after-hours support for unforeseen events such as the time when our customer TATA Motors had to spend 100 hours of manual labor to deploy, maintain, review, and analyze deep learning models. I worked to identify and develop a task allocation process on Nvidia servers that saved this manual work. I have a proven track record of consistently delivering exceptional service and support to clients.

Besides these engineering experiences and skills, I also possess interpersonal skills like being able to resolve unexpected and critical problems, particularly when working with diverse teams. For instance, at TCS, I played a key role in resolving issues with generalization in our prediction systems by presenting analysis reports and collaborating with researchers and managers from other core teams. Through effective communication, I helped to add intuitive features to optimize our ML models, resulting in reduced overfitting and improved prediction accuracy for critical events. While working to mitigate this issue, I developed hands-on skills in SQL, and Python, creating analytical dashboards and gaining a strong understanding of data science techniques and their applications.

I have demonstrated the ability to work in fast-paced environments. For instance, at Amazon, I worked to develop an early stopping mechanism in eligibility checks for enrolling vendors in the Amazon retail space. This improved the latency issues our team was facing before the holiday shopping season. I have experience in mentoring and collaborating effectively with budding associates. While at TCS, I worked with an intern and 2 new developers to document and study safety-critical scenarios in autonomous vehicles. We found that most practical methods in the literature were rule-based. However, we went above and beyond to develop deep learning models to learn more about high accident-prone scenarios. Furthermore, we leveraged the open-source UMich dataset (SPMD) to demonstrate the validity of our approach.