Product Engineer - NLP

• Oct 2021 - Present

Software Engineer

• Feb 2019 - Sep 2021

- Designed a framework to predict custom object labels for any dataset using multiple parallel pipelines to determine the best algorithm and parameter based on training score - Designed a framework to automatically annotate a custom dataset for image classification, semantic segmentation, object detection and recognition - Worked on object detection to create face detection and helmet detection models - Designed a framework to predict custom object labels for any dataset using multiple parallel pipelines to determine the best algorithm and parameter based on training score - Designed a framework to automatically annotate a custom dataset for image classification, semantic segmentation, object detection and recognition - Worked on object detection to create face detection and helmet detection models

Software Engineer

• May 2018 - Aug 2018

- Designed and implemented a backend API for the automation of chatbot creation for hotels - Improved the performance of intent classification by implementing multiple models (Word2Vec, fastText, LSTM) which helped in making the chat-bot more robust - Designed and implemented the database and backend API to build a food recommender system based on item-item collaborative filtering - Designed and implemented a backend API for the automation of chatbot creation for hotels - Improved the performance of intent classification by implementing multiple models (Word2Vec, fastText, LSTM) which helped in making the chat-bot more robust - Designed and implemented the database and backend API to build a food recommender system based on item-item collaborative filtering

Research Intern

• May 2017 - Jul 2017

Developed a new automatic clustering technique based on the fusion of multiobjective differential evolution and SOM that successfully beat some of the popular algorithms like VAMOSA and MOCK when used on text clustering Developed a new automatic clustering technique based on the fusion of multiobjective differential evolution and SOM that successfully beat some of the popular algorithms like VAMOSA and MOCK when used on text clustering

Senior Mentor for Autonomous Systems

• Aug 2016 - Jun 2017

- Review code and provide feedback to Udacity students on various Python, Machine Learning and Self-Driving Car projects. - Check code for function as well as performance as per Udacity's Style Guide - Provide feedback to Udacity team on student projects which includes clarification and support for student queries. - Evaluate around 30-40 projects a month with an average rating of 4.92/5. - Review code and provide feedback to Udacity students on various Python, Machine Learning and Self-Driving Car projects. - Check code for function as well as performance as per Udacity's Style Guide - Provide feedback to Udacity team on student projects which includes clarification and support for student queries. - Evaluate around 30-40 projects a month with an average rating of 4.92/5.

Open Source Community Member

• Jul 2016 - Jun 2017

DuckDuckGo is currently focusing on making it best search engine for programmers by providing instant answers and solutions to their queries. Instant Answers appear on DuckDuckGo.com above ads and organic links, on millions of searches. Instant Answers are created by an open source community of developers around the world. I have been working on: - Creating new Instant Answer modules using Perl and JavaScript - Fixing reported bugs in the existing Instant Answers -… Show more DuckDuckGo is currently focusing on making it best search engine for programmers by providing instant answers and solutions to their queries. Instant Answers appear on DuckDuckGo.com above ads and organic links, on millions of searches. Instant Answers are created by an open source community of developers around the world. I have been working on: - Creating new Instant Answer modules using Perl and JavaScript - Fixing reported bugs in the existing Instant Answers - Testing and reporting new bugs in existing Instant Answers - Code reviewing pull requests for new Instant Answers Show less DuckDuckGo is currently focusing on making it best search engine for programmers by providing instant answers and solutions to their queries. Instant Answers appear on DuckDuckGo.com above ads and organic links, on millions of searches. Instant Answers are created by an open source community of developers around the world. I have been working on: - Creating new Instant Answer modules using Perl and JavaScript - Fixing reported bugs in the existing Instant Answers -… Show more DuckDuckGo is currently focusing on making it best search engine for programmers by providing instant answers and solutions to their queries. Instant Answers appear on DuckDuckGo.com above ads and organic links, on millions of searches. Instant Answers are created by an open source community of developers around the world. I have been working on: - Creating new Instant Answer modules using Perl and JavaScript - Fixing reported bugs in the existing Instant Answers - Testing and reporting new bugs in existing Instant Answers - Code reviewing pull requests for new Instant Answers Show less