Dec 2023 – Apr 2024 · 4 mos · Hyderabad, Telangana, India · Remote

• Consultant

Jun 2023 – Aug 2023 · 2 mos · Hyderabad, Telangana, India · Remote

Jul 2022 – May 2023 · 10 mos · Hyderabad, Telangana, India · Remote

Jul 2021 – Jul 2022 · 1 yr · Hyderabad, Telangana, India

• Snapdragon Neural Processing Engine (SNPE) DSP/HTP Runtime
• ● Contributed to commercialization activities of SNPE by consistently debugging nightly and release issues.
• ● Implemented the new ONEHOT Op, LSTM Op, Detection Out Op for SNPE HTP resulting in enablement of 30+ models.
• ● Took charge to build a tool for debugging accuracy issues which automates the process of trying different quantization options with SNPE quantization tool. Thus saving 50+ developer hours per week by reduction of manual effort and optimization of the entire process.
• ● Tool for Recruitment Support Operations: - Initiated a side project across 10+ teams to create an automated one-stop for Resume Parsing for recruitment operations. - Automated filtering of resumes through extraction of personal information and skill set to save 10+ HR hours per week across 10+ teams. - This resume parser has now been adapted as the official recruitment tool for all the teams across the organization. - Tools & Technologies: Jupyter Notebook, Spacy, Python, Pandas, Bert, Huggingface, Flask, NLTK, Py-parser, Regex, Transfer Learning, Pytorch.
• ● Automated JIRA Handler: - Created framework to automatically assign JIRAS according to individual team structures and requirements. - Reduced manual effort by 70% and accelerated the process of issue management. - Tools & Technologies: Jupyter Notebook, Spacy, Python, Pandas, Bert, Huggingface, Flask, NLTK, Py-Parser, Regex, Transfer Learning, Pytorch.
• ● Proposed tool to optimize debugging flow by analyzing errors in logcat and linking similar issues.

Mar 2021 – Jun 2021 · 3 mos · Bangalore Urban, Karnataka, India

• ● Information Extraction from Video: - Developed framework hosted on AWS Elastic Beanstalk which analyzed audio and transcript from Zoom video interviews to extract information regarding technical questions asked. - Reduced additional cost of manual labor and human error by automating the process and achieving 90% accuracy score. - Tools & Technologies: Jupyter Notebook, Spacy, Python, Pandas, Bert, Huggingface, Flask, NLTK, Regex, Transfer Learning, Pytorch, AWS Elastic Beanstalk, YouTube Transcript Parser.
• ● Alert System for mismatch in Feedback: - Created a tool to detect anomalies between comments and scores in the interview feedback data provided. - Improved analysis time and decreased human error by 90% which resulted from QA manual process. - Tools Technologies: Jupyter Notebook, Spacy, Python, Pandas, Bert, Huggingface, Flask, NLTK, Regex, Transfer Learning, Pytorch, AWS Lambda, YouTube Transcript Parser.
• ● Question tag suggestion provider: - Built a tool to suggest various relevant tags for questions during QA analysis to provide insight on achieved scores. - Reduced effort required to manually find appropriate tags for questions. - Tools Technologies: Jupyter Notebook, Spacy, Python, Pandas, Bert, Huggingface, Flask, NLTK, Regex, Transfer Learning, Pytorch, AWS Lambda, YouTube Transcript Parser.

May 2020 – Jul 2020 · 2 mos · Kharagpur, West Bengal, India

• ● Developed Time Series Forecasting Model and Logistic Model to predict spread Covid-19 virus using Deep Learning techniques for short and long time periods respectively.
• ● Achieved accuracy of 98% for short time period of 15 days and of 81% for long time period of 365 days.

Jun 2018 – Jul 2018 · 1 mo · Jadavpur, kolkata

Dec 2017 – Jan 2018 · 1 mo · Kolkata, West Bengal, India