About

Sourav Kumar Goyal is a final year undergraduate student pursuing B.Tech in Instrumentation and Control Engineering at Dr. B R Ambedkar National Institute of Technology, Jalandhar. He is currently working as a Machine Learning Solution Engineer at Vernacular.ai. Sourav has a keen interest in Predictive Analysis, Data Visualization, Exploratory Data Analysis, and Building Machine Learning Models. He has hands-on experience in these areas, which can be seen on his LinkedIn profile. Sourav believes in adapting to new challenges, new spheres, and new cultures as a necessity to survive in this high-paced world. He has a natural instinct for analytical work and enjoys finding patterns, connections, and links within a system even in day-to-day life tasks. Sourav has chosen to become an Analyst as his professional career and is keen to find well-suited opportunities in the fields of Business Data Analytics, Machine Learning, and Data Science to kick start his career with a positive note on his professional life. Sourav has 1.86 years of relevant experience and has worked as a Data Science Intern at Crowdanalytix and Ownacar. At Crowdanalytix, he built a pipeline that automates the detection of colors in images to help retailers who formerly do this task manually with a higher degree of error. He also developed a Deep Learning Model to predict Primary & Secondary Color Family of the image using Image Data Augmentation and the Xception pre-trained model. At Ownacar, he built a pipeline that takes the image as input from the user, detects the car in that, and gives the probable output images of the car user is looking for. He also developed a Deep Leaning YOLOv3 Object Detection Model to detect the car in the input image and a Deep Learning Model that used a pre-trained model ResNet50 to extract features from the car images and uses Spotify's Annoy Search to recommend similar car images to that of the input. Sourav's Tech Stack includes ML, AI, Data Scientist, SQL, Deep Learning, NLP, Python, Scikit, Pandas, Computer Vision, Search, BERT, XGBoost, Object detection, Object Detection, test, and Frontend.