Aditya Mohanty r-2
Aditya Mohanty r-2
Aditya Mohanty r-2
Leetcode
GITHUB
SKILLS
Programming Language : Python , C++(Data Structures & Algorithms)
DS/ML/AI : ML, ANN/CNN Techniques , NLP ,BERT,Transformers
Python Libraries: Sklearn , Tensorflow ,nltk , keras , Numpy , Pandas ,Matplotlib , Seaborn,Langchain
Mlops Tools : Mlflow , git actions, DVC , CI/CD
Database language : Mysql
BI tool : Tableau
Other Technologies : Prompt Engineering ,LLM and Big Data , OOPS Concept
EDUCATION
Noida Institute of Engineering and Technology
2020- 2024
Bachelor of Technology
Computer Science (Specialization in Data Science) SCGPA : 7.5
URSULINE CONVENT SR. SEC. SCHOOL
AGGREGATE: 87.6% 2019 - 2020
Class - XII (CBSE)
Class - X (CBSE) AGGREGATE: 92.8% 2017 - 2018
CERTIFICATIONS
Machine Learning Essentials (AWS)
Data Visualization with Python (Coursera)
Exploratory Data Analysis -EDA(INFOSYS)
Natural Language Processing (DeepLearningAI)
WORK EXPERIENCE
VIRTUAL INTERNSHIP at KPMG (Data Analytics)
Created impactful data visualizations and reports using PowerBI and Tableau, resulting in a 20%
enhancement in data-driven decision-making.
Developed advanced techniques for extracting actionable insights from complex business datasets, leading to
a 15% improvement in decision-making accuracy.
Collaboratively executed all 3 crucial phases: data ingestion, transformation, and model training.
Authored a comprehensive presentation detailing the methodology for five key activities: data distributions,
feature engineering, data transformations, modeling, results interpretation, and reporting.
PROJECTS
1. MARKS PREDICTION (ML,CI/CD PIPELINES-Azure)
Collaboratively devised and optimized an advanced machine learning model, yielding a remarkable 87%
accuracy rate. Conducted thorough hyperparameter tuning for precision.
Seamlessly deployed the project using an Azure-based CI/CD pipeline, guaranteeing robust and rapid software
delivery.
Contributed to all project phases, following the Data Science workflows and other best practices for the project.
Upheld a dedicated commitment to Git for transparent communication, resulting in a 100% industry-ready
project.
2. MEDICAL SEARCH ENGINE (WORD2VEC and FAST-TEXT)
Designed and implemented a medical search engine employing trained Word2Vec and Fast-Text models,
achieving a noteworthy 81% accuracy rate.
Identified the necessity for domain-specific embeddings, culminating in the creation of medical word
embeddings that significantly improved search accuracy.
Managed a dataset comprising 10,666 rows and 21 columns, facilitating efficient medical information retrieval.