About

Dilip Upadhyay is an Associate Technical Consultant at KPMG with more than 2 years of experience in Data Engineering and Data Science. He has a strong background in machine learning use cases, including pyspark use case, and is skilled in data cleaning, feature engineering, training different models, and creating inference services over flask. Dilip has also gained experience in MLOPS, Data Science, and Machine learning algorithms. Dilip is proficient in programming skills, with Python being his language of choice. He has worked with various tools such as HDFS, Pyspark, Mllib, Flask, Docker, Linux, Shell Scripting, Tensorflow data validation (tfdv), Tensorflow model validation (tfmv), AWS SageMaker, and Azure ML. In his previous role as an Associate Software Engineer at Abzooba, Dilip worked on a banking solution with data having more than 80M rows. The end goal was to predict whether a loan applicant would pay the loan or not using the Random Forest algorithm, vector assembler, and string indexer. He also worked on an apparel attribute classifier use case, where he created a flask-based inference service that predicted the apparel attribute (sleeve, color, text) from the image. Dilip's role included creating docker images for the inference service code and saved models. As a Software Engineering Trainee at Abzooba, Dilip worked on an MLOPS product (Xpresso.ai) development team and created a part of the product, model monitoring. The end goal of the product was to monitor the decay in models after being released in production. He also worked with AWS SageMaker and Azure ML for comparing the feature availability with respect to Xpresso.ai. Dilip holds a Bachelor of Technology in Computer Science from Haldia Institute of Technology. He is skilled in Flask, Data Engineering, ML, Data Scientist, Python, Docker, Random Forest, HDFS, Software Engineer, AI, Tensorflow, AWS, Sagemaker, Azure, and has relevant experience of 3.87 years.