Hello World. This is
Raunaq Jain
Your friendly neighbourhood Software Developer
I'm a Software Development Engineer at Audible, with a mission to make our customers have the best experience listening Audiobooks. I graduated from the University at Buffalo with a Masters in Computer Science and an immense desire to create something life-changing.
I have been involved in the domain of Machine Learning since my undergrad, and even have a publication against my name!
Outside of work, I am either playing Xbox, or watching movies, anime or F1.
About Me
Hello! I'm Raunaq, a Software Developer at Audible. I enjoy solving challenges and building applications for a purpose. My goal is to develop exceptional features that support, brighten and entertain users in their daily life.
I completed my Masters in Computer Science from the University at Buffalo; wherein I worked with Prof. Varun Chandola & Prof. Marc Böhlen to develop an application that classifies images and videos of plant species in wild and publish our results - "From images in the wild to video-informed image classification", 20th IEEE ICMLA, 2021 link.
Before that, I interned at Hypothizer, where I built a cloud-native document processing platform to capture intelligence-ready data.
Here are a few technologies I've been working with recently:
- Python
- Java
- Rust
- Django
- AWS
Where I've Worked
Software Development Engineer @ Audible
April 2021 - Present
- Lead the migration of 950+ hosts to Amazon Linux 2 for enhanced performance and security benefits.
- Developed an Integration Test package for Audible Rights Service which currently serves 500 requests of batches per second.
- Improved team’s agile processes as a scrum master and part of the on-call rotation triaging tickets.
- Pipeline Ambassador: Responsible for monitoring the health of 50+ pipelines and lead them towards full CI/CD.
Volunteer @ AI Ethnobotany - University at Buffalo
June 2020 - Present
- Developing an application in Flask to process and classify images and videos of plants in the field with Prof. Varun Chandola & Prof. Marc Bohlen.
- Training a Single Shot Multi-Box Detector to detect plants.
Volunteer @ RMS Lab - University at Buffalo
June 2020 - December 2020
- Designed a secure file storage system by implementing a hash tree in Rust for Trusted Execution Environment (TEE) under Prof. Steven Ko.
- Performed software verification to verify absence of memory leaks and dangling pointers in the kernel.
Machine Learning Intern @ Nulenta
December 2019 - January 2020
- Spearheaded development of a resume parser in Django (REST API).
- Built an end-to-end solution to match and score job requirements with user profiles for automatic candidate selection using a machine learning model.
Machine Learning Research Intern @ Hypothizer
July 2018 - February 2019
- Contributed extensively to a cloud-native document processing platform for enterprises that captures intelligence-ready data from documents.
- Improved accuracy of existing invoice parser by 10% through feature engineering and custom machine learning models.
- Architected, developed, and shipped Resume Parser with an average F1 score of 0.93 after training on 150 documents.
- Extensively researched various deep learning approaches and implemented image processing algorithms.
Research and Development Intern' @ All India Council for Technical Education
October 2017 - March 2018
- Created a web application in Django and utilized machine learning to predict employment potential of universities and set standards for 10k+ technical colleges throughout India. Updated AICTE’s Approval Handbook. Documentation
- Developed data pipeline and trained machine-learning models through scikit-learn.
- Curated, analyzed, and visualized data of 5 years in Python and developed interactive plots in Plotly for user interface.
- Awarded with the 1st prize ($3000) in Smart India Hackathon’17, MHRD, Govt. of India, against 7400+ teams.
Some Things I've Built
Featured Project
DogDogGo
An end-to-end tweet search engine indexing 200K multi-lingual tweets using Apache Solr. Impact of political rhetoric in traditional and social media is analyzed and presented using interactive geospatial plots. Includes features like tf-idf ranking, pseudo relevance feedback, more like this, filter-based search, query translation, and search highlighting.
- Django
- React
- Plotly
- Apache Solr
Featured Project
FEVER: Fact extraction and Claim Verification
A web application to check the credibility an input claim. Relevant documents are fetched from Wikipedia 2017 data dump and evidential sentences are selected from amongst them. The claim is classified into SUPPORTED, REFUTED, or NON-ENOUGH INFO based on the evidences.
- Flask
- NLP
- Apache Solr
- Nginx
Other Noteworthy Projects
view the archivePolicy Gradient Methods (Reinforcement Learning)
Implemented REINFORCE and Advantage Actor-Critic algorithms to analyze their advantages and disadvantages.
Atari Breakout (Reinforcement Learning)
Trained an agent to play Atari Breakout achieving state-of-the-art results.
Computer Vision
Implementation of image stitching, kmeans segmentation, image denoising using filters, and ransac algorithms.
Similarity Between Us - Siamese Networks
A face recognition system inspired from Google's FaceNet paper.
What's Next?
Get In Touch
I'm currently looking for job opportunities, my inbox is always open. Whether you have a question or just want to say hi, I'll get back to you!