Prajwal Krishna
Aspiring Data Scientist
Real world capabilities
I specialize in crafting and executing robust ML infrastructure and automation pipelines, harnessing cutting-edge tools and frameworks like Docker, Kubernetes, Git Actions, Huggingface models, and more to streamline workflows, ensuring scalability and reproducibility of ML processes. My proficiency in seamlessly integrating ML solutions into existing systems empowers organizations to fully leverage their data, gaining a competitive advantage in today’s dynamic market landscape. If you’re in search of a skilled ML/DL engineer and MLOps expert capable of delivering exceptional results, I am confident in my ability to propel your organization’s success through innovative technologies and data-driven solutions. Let’s collaborate and unlock the potential of ML and DL for your business together.
You can find my works here
-
Omdena-Inc - Machine Learning Engineer
Mar 2024 - Ongoing- The project's core objective is to advance sustainable business practices through the creation of an AI-driven tool tailored for automating climate disclosure reporting.
- This endeavor is pivotal in facilitating businesses to effectively fulfill regulatory obligations by harnessing the power of AI and large language models (LLMs) to simplify the drafting process of reports.
-
iNeuron - Data Scientist | Intern
Jan 18 2023 - Feb 16 2024- Spearheaded an End-to-end MLOps Predictive Maintenance Project, achieving 98%+ performance across two classification models.
- Leveraged python libraries (pandas, numpy, sklearn) for data manipulation and analysis. Implemented Mlflow for tracking and experimentation, Utilized Evidently AI to test and monitor ML models in production.
- Expertly deployed models utilizing Docker for containerization, Streamlit to develop front end application and Github actions for version control.
- Ensured data quality and relevance through expert cleaning, transformation, thorough Exploratory Data Analysis, and Model Monitoring.
-
BCG-Forage - Data Science Internship
Apr 2023 - Jun 2023- Executed a customer churn analysis for PowerCo Analytics by demonstrating advanced data analytics skills, identifying essential client data and outlining a strategic approach that enhanced client retention strategies.
- Conducted efficient data analysis using Python, including Pandas and NumPy. Employed data visualization techniques for insightful trend interpretation.
- Engineered and optimized a random forest model, achieving an 85% accuracy rate in predicting customer churn, significantly enhancing predictive accuracy.
-
Pantech-Solutions - Machine Learning Intern (trainee)
Nov 2022 - Jan 2023- Learn ML & DL from Scratch
- Learnt to build neural networks using ANN,CNN,RNN,LSTM etc
- Build numerous types of projects by using different ML and DL algorithms
-
App-Brewery - Python trainee
Jun 2022 - Jan 2023- Learn Python from Scratch.
- Learnt to build fuly fledged Websites and Web Apps with python.
- Built numerous types of projects, games, apps and automation.
- The project's core objective is to advance sustainable business practices through the creation of an AI-driven tool tailored for automating climate disclosure reporting.
- This endeavor is pivotal in facilitating businesses to effectively fulfill regulatory obligations by harnessing the power of AI and large language models (LLMs) to simplify the drafting process of reports.
- Spearheaded an End-to-end MLOps Predictive Maintenance Project, achieving 98%+ performance across two classification models.
- Leveraged python libraries (pandas, numpy, sklearn) for data manipulation and analysis. Implemented Mlflow for tracking and experimentation, Utilized Evidently AI to test and monitor ML models in production.
- Expertly deployed models utilizing Docker for containerization, Streamlit to develop front end application and Github actions for version control.
- Ensured data quality and relevance through expert cleaning, transformation, thorough Exploratory Data Analysis, and Model Monitoring.
- Executed a customer churn analysis for PowerCo Analytics by demonstrating advanced data analytics skills, identifying essential client data and outlining a strategic approach that enhanced client retention strategies.
- Conducted efficient data analysis using Python, including Pandas and NumPy. Employed data visualization techniques for insightful trend interpretation.
- Engineered and optimized a random forest model, achieving an 85% accuracy rate in predicting customer churn, significantly enhancing predictive accuracy.
- Learn ML & DL from Scratch
- Learnt to build neural networks using ANN,CNN,RNN,LSTM etc
- Build numerous types of projects by using different ML and DL algorithms
- Learn Python from Scratch.
- Learnt to build fuly fledged Websites and Web Apps with python.
- Built numerous types of projects, games, apps and automation.
-
Full-Stack-Data-Science-Masters - iNeuron.ai
Apr 2023 - Ongoing -
Tensorflow-Developer-Certificate - Deeplearning.ai
Jul 2023 - Sep 2023 -
Deep-Learning-Specialization - DeepLearning.ai
Jul 2023 - Aug 2023 -
Machine-Learning-Specialization - DeepLearning.ai
Jun 2023 - Aug 2023 -
Mathematics-for-Machine-Learning-and-Data-Science - DeepLearning.ai
Jun 2023 - Aug 2023 -
Python-for-Data-Science-and-Machine-Learning-Bootcamp - Pierian Training
Aug 2022 - Nov 2022
-
MCA (Artificial Intelligence and Data Science)
Visvesvaraya Technological University2023 - Present -
Bachelor of Science (Aviation)
Singhania University2019 - 2022 -
Class XII-
SSEA Pre University College2018 - 2019
Predictive Maintenance
Predictive maintenance helps companies minimize downtime, reduce repair costs, and improve operational efficiency. Developing a web application for predictive maintenance can provide users with real-time insights into equipment performance, enabling proactive maintenance, and reducing unplanned downtime.
Onsite Health Diagnostics-OHD
Onsite Health Diagnostic (OHD) is the web application that allows users to predict whether the user has been infected with a menacing disease or not. These diseases can be very dangerous to health if they are not treated properly. The main objective of OHD is to help people predict the disease in case of absence of medical professionals, strikes or any related uncertainties.
Jigsaw - Multilingual Toxic Comment Classification
The goal is to find the probability that a comment is toxic. Built a multilingual text classification model to predict the probability that a comment is toxic based on 4,35,775 text comments in 7 different languages, using the data provided by Google Jigsaw. Fine-tuned BERT-Multilingual-base and XLMRoBERTa models on the multilingual text.
Airline Passenger Satisfaction Prediction
Following the pandemic, the airline industry suffered a massive setback, with ICAO estimating a 371 billion dollar loss in 2020, and a 329 billion dollar loss with reduced seat capacity. As a result, in order to revitalise the industry in the face of the current recession, it is absolutely necessary to understand the customer pain points and improve their satisfaction with the services provided.
Realtime Document Chat System
The project aims to develop a PDF querying system that leverages LangChain, a powerful language processing tool, to extract information from PDF documents. By employing LangChain's advanced natural language understanding capabilities, the system will enable users to perform complex searches and obtain specific data points from PDF files efficiently and accurately.
Consignment Pricing Prediction MLOPS
Designed and built an end-to-end scalable machine learning pipeline using Mlops in which i have integrated everything on a frontend so that user can get log file,artifacts,model binary file and also save the data to the database which can help for future retraining and also you can perform complete retraining of the model by just one click.The main aim is to build an entire system of a machine learning so that it can be used in realworld scenario.
Insurance Premium Prediction - Mlops
The End-goal of this project is to give people an estimate of how much they need based on their individual health situation.So here I have implemented End-End Machine Learning project using MLOPS lifecycle.Built CI/CD pipeline.DVC(Data Version Control) was used here for track ml model.It was regression problem statement and used GradientBoostingRegressor to solve the problem which gave r2_score of 85% after tuning parameters.
Automated Machine Learning Application
The Automated ML WebApp project is a tool developed using Python, Pandas Profiling, PyCaret, and Streamlit frameworks. It aims to simplify and streamline the process of building machine learning models by automating various tasks involved in data preprocessing, feature engineering, model selection, and downloading the trained models.