Lakshmi Vadhanie

Graduate Teaching Assistant @ Northeastern University

About

Hi, I’m Lakshmi Vadhanie, I'm currently pursuing my Master’s in Data Science at Northeastern University. I’m passionate about leveraging AI, machine learning, and data analytics to solve real-world problems and create meaningful impact, whether it’s improving healthcare outcomes, enhancing financial systems, or helping organizations make smarter, data-driven decisions. I love tackling complex problems and turning them into actionable insights. My work spans building personalized treatment recommendation systems using BERT, securing machine learning models against data poisoning, and developing AI-powered platforms for real-time risk detection. Each project has strengthened my ability to bridge technical expertise with strategic thinking. A core goal for me is to develop solutions that are not only precise but also truly contribute to positive outcomes and real-world impact. I’ve also published my research in peer-reviewed journals: - personalized treatment recommendations (Book Chapter) - algorithmic resilience against data poisoning (Journal, won first prize, Malaysia) This experience gave me invaluable experience in communicating complex ideas clearly and collaborating with diverse teams. I thrive in environments that encourage curiosity, experimentation, and continuous learning. I enjoy collaborating with cross-functional teams, translating data into compelling insights, and seeing my work drive tangible outcomes. Technically, I am proficient in Python, SQL, AWS, PyTorch, TensorFlow, Spark, LLMs, and visualization tools like Tableau and Power BI. I’m always excited to connect with people who are passionate about using data to drive innovation and make a real difference. Feel free to reach out to me at ganesh.la@northeastern.edu, I’d love to chat!

Country

United States

City

Boston

Industry

Information Technology & Services

Skill

Amazon Web Services (AWS), Cloud Computing, AWS Identity and Access Management (AWS IAM), AWS SageMaker, Computer Vision, Natural Language Processing (NLP), Neural Networks, Computer Science, Linux, Red Hat Linux, Cascading Style Sheets (CSS), HTML5, Deep Learning, Artificial Intelligence (AI), Communication, Machine Learning, Data Science, Data Analytics, Data mining, Data Structures

Experience

Graduate Teaching Assistant

Northeastern University

2026-1 - Present · 9 mos

Boston, Massachusetts, United States

CS 7180: SpTp AI-Assisted Software Engineering

Machine Learning Co-op

Wave Life Sciences

2025-7 - 2025-12 · 6 mos

Cambridge, Massachusetts, United States

Research Assistant

SRM Institute of Science and Technology (SRMIST)

2023-6 - 2024-5 · 1 yr

• Developed a personalized treatment recommendation system using BERT-based model, achieving 92% accuracy in classifying drug recommendations through fine-tuned embeddings and precision-focused metrics (precision, recall, F1-score). • Leveraged transfer learning with pre-trained BERT model to enhance sentiment classification using tokenization, attention mechanisms, and hyperparameter tuning. • Enhanced CNN security by designing systems to detect and mitigate data poisoning attacks, improving model accuracy from 78% to 91% using data augmentation, adversarial training, and ensemble learning. • Optimized model performance using Gaussian processes and Bayesian optimization, boosting CNN resilience by 10% with adversarial defense techniques like gradient masking and input filtering

AWS AI-ML Intern

All India Council for Technical Education (AICTE)

2023-9 - 2023-11 · 3 mos

India

- Streamlined identity and access management (IAM), VPC, and web server deployment in AWS. - Implemented ML pipeline using SageMaker- Computer vision and NLP techniques.

Intern-Machine Learning Engineer

Zoho

2023-6 - 2023-7 · 2 mos

Chennai, Tamil Nadu, India

• Developed production ML systems including an LSTM-based forecasting pipeline (+15% over ARIMA) and LLM evaluation workflows using perplexity, ROUGE, METEOR, and BLEU (+20% quality gain) • Reduced IT ticket resolution time by 45% by deploying an agentic support system with LangChain, GPT-4, Pinecone, and FastAPI, improving response accuracy from 67% to 89% at 500+ queries/day

Education

Northeastern University

Data Science

2024 - 2026 · 2 yrs

Relevant coursework: DS5220 Supervised Machine Learning CS5800 Algorithms CS6220 Data Mining Techniques DS5983 Large Language Models (LLMs) DS5110 Database Management and Processing IE7374 Machine Learning Operations (MLOps)