Ambigapathi V Email & Phone Number

As an aspiring Data Scientist with a strong foundation in agriculture, I am passionate about harnessing data to drive insights and informed decision-making. My journey in data science has equipped me with a diverse skill set, including proficiency in Machine Learning, Generative AI, and Natural Language Processing (NLP). I specialize in developing predictive models and leveraging data visualization tools to present complex information clearly and effectively.I possess strong programming skills in Python and SQL, and I utilize libraries such as Pandas and NumPy for data manipulation. My expertise in data visualization tools like Power BI allows me to present insights that drive business outcomes. Additionally, I have experience with FastAPI for deploying applications and am familiar with deep learning frameworks like TensorFlow and PyTorch.I am committed to continuous learning and staying abreast of the latest advancements in data science. With strong communication and collaboration skills, I am eager to contribute to innovative projects and work alongside like-minded professionals in the field. Let’s connect and explore opportunities to leverage data for impactful solutions!

• 

  Machine Learning Engineer

  Omdena

  Sep 2024 - Present - Kitwe, Copperbelt Province, Zambia

  Project name: local news aggregator platform for Kitwe, ZambiaAs part of Omdena’s global team, I contributed to developing a local news aggregator platform for Kitwe, Zambia. The project aimed to combat misinformation and provide actionable insights to the community by leveraging advanced machine learning and NLP techniques.Key Contributions:1.Data Collection and Credibility Analysis: Aggregated news from diverse sources and implemented a system to evaluate source credibility, classifying them as reputable, suspicious, or unknown.2.Fake News Detection Features: Designed features like clickbait detection, sentiment analysis, keyword frequency analysis, and topic modeling with LDA to identify sensational and misleading content.3.Model Development: Fine-tuned transformer models, including BERT, DistilBERT, and RoBERTa, selecting BERT-base-uncased for its superior performance in accuracy and efficiency.4.Class Imbalance Mitigation: Applied SMOTE to address dataset imbalances, improving prediction robustness.5.Streamlit Dashboard: Built and deployed a user-friendly Streamlit app, enabling real-time news insights and classification, hosted on Streamlit Cloud.6.Evaluation and Testing: Ensured model reliability using metrics such as accuracy, F1 score, and AUC-ROC, while optimizing the platform for usability and efficiency.This project strengthened my skills in NLP, transformer-based models, data engineering, and app development, while contributing to a platform that leverages technology to foster informed decision-making and social impact.

• 

  Annamalai University

  Agriculture