Experience

---

July 2021 – Oct 2022

Associate Data Scientist

Built a context-aware chatbot using Rasa. The bot handles unexpected behavior and maintains context. Brought improvements in related areas like entity extraction and user intent detection. Worked with models like Conditional Random Fields and transformer-based architectures.

I also designed packages to automatically generate mock data for the initial training. Based on a series of templates and entities from our data sources, the generated data was good enough to allow the bot to converse well in most scenarios.

Jan 2021 – July 2021

Data Science Intern

Built a multi-label text classifier using random forests to categorize
descriptions of services done on vehicles.
Achieved 3x results on metrics like recall, f1-score, and coverage while
maintaining the same high precision.

Filed a patent for my work on efficiently generating textual training data and its integration into a service recommendation model.

May 2020 – July 2020

Machine Learning Intern

Built a dynamic model for anomaly detection in RAN KPIs which detects
anomalies in real-time using raw data.
Achieved an accuracy of over 90% for some critical KPIs.

May 2020 – July 2020

Open-source Mentor @CHAOSS

I mentored Venu Vardhan Reddy Tekula on his project for Google Summer of Code’20 with CHAOSS.

The project involved building Quality Models for various metrics, making assessing the health of open-source projects and communities more accessible. As a mentor, my responsibilities included providing suggestions and feedback and guiding Venu over the course of the project.

May 2019 – august 2019

Open-source Developer @CHAOSS

I was accepted as a Google Summer of Code student for CHAOSS. My project involved analyzing data fetched via Perceval and creating reference implementations for CHAOSS metrics. I extensively used Pandas, Matplotlib, and Jupyter Notebooks. More information about my project can be found here.

Aug 2020 – mar 2021

Undergraduate Researcher

I worked under Dr Amit Dua and Murari Mandal to obtain significantly improved results on medical radiology image classification. The focus of the study was on the Stanford MURA dataset.

The dataset was part of an online competition held by Stanford University. The study tests conventional techniques like vanilla CNNs and compares them to the performance of the state-of-the-art model by Stanford. It also discusses a few other approaches which I tried but were not significant and did not perform as well as I had initially hoped.

The study examines another technique which uses the concept of Attention. Finally, the project attempts another approach by combining a DenseNet with a Hierarchical Attention Network and then concludes by suggesting potential strategies which may give promising results.

Worked on de-raining images using GANs.

---

Jan 2019 – Jun 2019

Teaching Assistant

Worked as a teaching assistant for the introductory programming course at Bits Pilani, which covers the basics of C and the Unix command line.

As a teaching assistant, I conducted weekly lab sessions and a once-a-week doubt-clearing session with my students. Along with furthering the depth of my understanding of the C language, this opportunity made me a better and more confident communicator.

Jun 2018 – Jul 2018

Machine Learning Intern

My first internship. I tested several models for denoising and deblurring images of printed text. I also improved the parsing of addresses using Conditional Random Fields and OpenStreetMap data.

Feb 2018 – Mar 2019

Backend Web Dev and ML Enthusiast

I was part of the backend web development team as well as the Machine Learning team. I worked to design fest websites with Django and conducted a backend web development workshop with my team.

As part of the ML team, I worked on foundational projects that improved my understanding of the fundamentals. We also conducted workshops and other fun events for fests.

Aug 2017 – Mar 2019

Backend web-dev and ML Enthusiast

I became a member of this club in my first year at BITS Pilani. I was a backend web developer for a while, but then also joined the ML team. I worked on the backend part of several websites, especially those of fests and other major events. In my second year, I helped the new first-year recruits get a good grasp of Django, the Unix command line and git.

The ML team organized a hackathon and encouraged its members to pursue team and individual projects.