Apr 2023 – Dec 2023 · 8 mos

Oct 2021 – Mar 2023 · 1 yr 5 mos

• Projects completed for building micro-services for lightweight models and platforms. Working towards building SDKs and libraries of product for 3rd parties to integrate Wellthy's digital therapeutics system into their platforms.
• Technical Details:
• Integrating optimised CI/CD systems in Jenkins to automatically build models and servers from Github repositories.
• Developing scalable microservices and working towards reducing overall system resources of servers.
• Building SDK platforms and on device data management systems for securely integrating Wellthy's product in 3rd party systems without compromising on data.

Apr 2019 – Sep 2021 · 2 yrs 5 mos

• Projects taken up with end to end responsibility for architecture, development and managing the company's digital insight system and patient chat platform.
• Technical Details:
• Developed scalable patient and health mentor chat platforms serving thousands of patients daily under low bandwidth constraints.
• Designed and developed patient insight system for digital insights regarding meal, activity and other parameters based on patients' and suggesting health goals and better living standards.
• Deployment and maintenance of main platform systems on AWS and auto scaling server architecture.

May 2019 – Oct 2020 · 1 yr 5 mos · Remote

• Working as an AI consultant remotely to a project for building better detection algorithms as a part of Berkeley SETI Research Center. Project under SETI (Search for Extra-Terrestrial Intelligence) and UC Berekely SETI Research Center.
• Technical Details:
• Signal processing for bifurcating anomalies.
• Image processing for understanding Radio waves behavior using signal wave images processed through deep convolution neural networks.
• Scalable model development for real time inference building.
• Multi class signal classifiers for detailed analysis of radio waves.

Mar 2018 – Mar 2019 · 1 yr · Bengaluru Area, India

• Projects taken as a part of early stage AI startup for Fintech domain and building solutions end to end. AI solutions given out over multiple channels and APIs.
• Tasked with development of cross platform highly scalable and high precision NLU for Fintech.
• Technical Details:
• Built Deep Neural Networks for Natural Language Processing tasks like Intent Detection, detailed sentiment analysis, named-entity recognition, word embedding models, sense2vec development and natural query parsing.
• Asynchronous Speech recognition with help of deep neural nets for speech analysis and intent detection on speech.
• Computer Vision as a part of document parsing and data extraction and identification from images.

Jan 2018 – Jun 2018 · 5 mos · Bangalore

• Tasked with development of automated frameworks that help with continuous integration of projects.
• Techincal Details:
• Code flow analysis for predicting code coverage based on provided test cases.
• Prediction models for predicting bugs in a code and high-risk code areas for runtime bugs.
• Framework for doing automated build and unit tests for continuous integration.

Sep 2017 – Oct 2017 · 1 mo · Bengaluru Area, India

• Tasked with building a platform that uses employee profiles and uses data extraction on natural language profile descriptions for better internal affairs. Project done for an industry leading organization with large workforce and international presence. Project undertaken as a part of a small team to develop the models and ship them as a web application.
• Technical Details:
• Natural Language Processing used for multiple tasks like Data Extraction, named-entity recognition and profile analysis.
• Neural networks and NLU frameworks like RASA used for natural language query processing and drawing intents and insights from user queries.
• Machine Learning algorithms like Resnik similarity between words in an embedding model used for determining semantic similarity between 2 sentences.

Jun 2017 – Jul 2017 · 1 mo · Bengaluru Area, India

• Worked as a part of a team developing new multipurpose naval radar for Indian defense forces. Project undertaken as a part of BEL software Technology Center.
• Technical details:
• Developed a framework for linking RADAR data with onboard computers.
• Visualization of data for GUI based RADAR interface using Qt.
• Machine Critical Algorithms for instant analysis.

May 2017 – Sep 2017 · 4 mos · Bangalore

• Worked on analyzing high amounts of data generated by sensors and predictive analysis of failures. Building lightweight models to be deployed on embedded systems. Projects was taken as a part of Embedded Systems team under Boeing India.
• Traditional Machine Learning systems for developing scalable and lightweight models for predictive analysis.
• Fault tolerant systems development with 3 layers of redundancy.
• Real time data analysis and insights.