Master’s in Computer Science, seeking to apply analytical and programming expertise within diverse IT roles. Spe cializes in Python, AI, software development, and machine learning applications. Committed to utilizing technology to drive efficiencies and advance IT solutions in dynamic environments. Seeking to leverage my technical skills and passion for problem-solving in a dynamic and challenging environment.
As a Master Thesis Student I am currently working on providing Actionable insights on ITSM Incident Management data, by building a LLM chat application using LangChain and OpenAI that can provide the insights for an incident selected.
As a Research Intern conducted research on prediction of ADMET properties. Where I have done literature surveys to explore the prior research on these properties. Developed RF and XGB model to predict Extraction (E) property
Conducted an Exploratory Data Analysis (EDA) on the 'Global Terrorism' dataset using Python to identify terrorism hot zones and derive key security insights which involved data cleaning, trend analysis and tactics of terrorist attacks.
I have a robust foundation in computer science, my comprehensive coursework includes applied artificial intelligence and machine learning, equipping me with advanced skills in developing intelligent systems. I have hands-on experience in programming within UNIX environments and a solid understanding of decision support systems.
Earned a Bachelor of Technology degree in Computer Science encompassing a comprehensive curriculum that covered essential principles of computer science such as algorithms, data structures, machine learning, software development, computer networks, operating systems, and database management, which effectively help me in doing my internship.
Developed an innovative POC RAG application leveraging Langchain, Chromadb, and Ollama to answer com plex questions related to IT Incidents Root Cause Analysis (RCA) data. This application streamlined incidents, providing actionable insights that enhanced incident resolution efficiency.
Engineered an advanced AI chatbot for an e-commerce coffee platform using PyTorch and natural language processing (NLP) techniques. Leveraged a sophisticated feed-forward neural network with two hidden layers.
Designed and implemented an ensemble machine learning model by integrating SVM, and Naive Bayes algo rithms and compared with RF and XGB. This comparison led to a significant boost in language detection accuracy, reaching an impressive precision 96% and recall 91% it tends to accurately predict the correct lan guage classes while also being effective at capturing most relevant instances than other models.
Deployed a cutting-edge real-time credit card fraud detection system, leveraging Flask for the backend and advanced machine learning algorithms. Achieved 93% accuracy in identifying fraudulent transactions, signifi cantly bolstering transaction security and mitigating financial risks.
Developed a sophisticated decision support system utilizing Python and MongoDB to identify the optimal machine learning model for short-term cryptocurrency price prediction. Evaluated multiple models, including GRU, LSTM, Bi-LSTM, XGBoost, and Random Forest, to ensure comprehensive analysis and selection. This innovative approach enhanced prediction accuracy and provided critical insights for informed decision-making in the volatile cryptocurrency market.
Implemented robust models to detect and flag inappropriate or harmful online comments. Leveraged advanced natural language processing (NLP) techniques to pre-process and extract critical features from text data. Utilized these features to train a highly accurate classification model, significantly improving the identification and mitigation of toxic content, thereby fostering a safer and more respectful online environment.
The project focuses on allocating cloud resources to servers using Partition-based load balancing algorithm and uses swing API for GUI for visualization of cloud activities.
The project’s main idea is to detect one or more drones appearing at some point in a video sequence where birds and other distracted objects may also be present, together with motion in the background done by image classification using CNN.
The project’s main idea is to detect one or more drones appearing at some point in a video sequence where birds and other distracted objects may also be present, together with motion in the background done by image classification using CNN.
