I embarked on my programming journey during my academic years at IIT (ISM), Dhanbad. My favorite part of programming is the problem-solving aspect.
With over 2 years of current experience as a Software Engineer, I started my professional journey as a Machine Learning Engineer where I ventured into Computer Vision, Classical ML and Image Super-Resolution, where I've honed my skills with Generative Adversarial Network, Decision Trees models using TensorFlow, Python and C to classify cattle activity using signals obtained from 3D Accelerometer. Going forward, I got extensive experience as a Backend Developer in cutting-edge technologies like Golang , gRPC , Protocol Buffers and Grule engine which I've leveraged to build a recommendation system to generate dynamic advisories. Also, got opportunity to develop and optimize Java Spring Boot APIs for a SIEM platform, improving performance through Redis Caching and database indexing. Utilized Test-Driven Development (TDD) to achieve 90% code coverage and wrote comprehensive E2E integration tests. Implemented robust monitoring and error logging using SLF4J.
Beyond my professional endeavors, I've undertaken personal projects utilizing TypeScript, Node.js, Next.js, React.js and MongoDB for Web Dev and leveraged OpenAI API, further solidifying my versatility and dedication to mastering new technologies. Also, during my college years, I dedicated some time to Android Dev, utilizing Java to build basic applications.
Implemented Java SpringBoot APIs for SIEM platform where we manage different assets and for every asset based on user action will generate log lines which needs to be monitored as they might contain potential security breaches and if thats the case, we throw alerts.
Security Information and Event ManagementImplemented Golang based Grule Engine with structured Protocol Buffers schemas utilizing gRPC for building communication channels to generate dynamic advisories for farmers from diverse datasets of soil health, crop types and geographic locations.
AgriCare Fertlizer Recommendation SystemImplemented Generative Adversarial Network(ESRGAN) model using Python TensorFlow for enhancing image resolution to 4 times.
Satellite Imagery Resolution EnhancementDeveloped a backend service using Python automating downloads of image tiles from QGIS(Quantum Geographic Information System) at a particular resolution for given diagonal Lat/Lng and stored in Azure Blob.
Automated Image Download and Cloud Storage IntegrationUtilized R-tree Data Structure for accelerating Nearest Neighbor Searches based on Lat/Lng reducing search time from 7.5 secs to 0.5 sec.
Lightening Fast Geospatial QueriesDeveloped classification model for real-time monitoring of cattle behavior using Python utilizing GridSearchCV for hyperparameter tuning to optimize model performance and translated code to C for deployment on embedded system
Cattle Activity Classification ModellingDeveloped responsive Question and Answers platform where users can post queries with associated tags, receive community-driven responses, personalized recommendations and integrated state-of-art Open AI API's to generate AI answers.
Developed a places listing web app where user can browse and book places with intuitive calendars based on there availability for that particular time period and integrated Amozon S3 as cloud storage solution
Developed an Android application where users can find friends, send requests, chat and exchanges multi-media messages
Please reach out to me directly at abhay5349singh@gmail.com or through this form.