M.Sc. Data Science @ VIT-AP · IEEE Published Researcher · TCS Digital (AI Cloud) Selectee
I build end-to-end data products - from cleaning messy inputs and engineering features to training models, designing analytics, and shipping usable interfaces. My work sits at the intersection of machine learning, deep learning, knowledge graphs, and decision-focused analytics.
abhishek = {
"role": "AI-Powered Data Insights Intern @ TOEHO AI",
"building": ["ML platforms", "knowledge graphs", "analytics dashboards"],
"exploring": ["LLMs", "RAG pipelines", "production ML"],
"open_to": "Data Science and ML Engineering opportunities",
}
Watch the agent collect data, learn representations, train a model, and deliver a prediction.
- AI-powered data insights: extracting and analyzing signals from social media and news sources at TOEHO AI.
- No-code machine learning: making data cleaning, feature engineering, visualization, and model training more accessible.
- Connected knowledge: combining web crawling, graph architecture, and relevance ranking for guided learning.
- Applied research: studying how learning rates, optimizers, and activation functions interact during CNN training.
🕸️ Graph NetAn intelligent knowledge-graph application that combines web search, NLP, Gemini-powered summaries, and interactive concept relationships.
|
A no-code data science platform with drag-and-drop workflows for preprocessing, visualization, feature engineering, and model training.
|
|
A graph-based generative system using a variational autoencoder and relational graph convolutions to generate valid drug-like molecules.
|
A patient-centered booking experience redesigned around clearer navigation, stronger calls to action, and improved accessibility.
|
Investigating the Synergistic Effects of Learning Rate, Optimizer, and Activation Function on Convolutional Neural Network Training
Published at the 2025 IEEE 5th International Conference on Artificial Intelligence and Signal Processing. The work compares optimizer and activation-function combinations with dynamic learning-rate tuning to study CNN convergence, stability, and generalization.
Explore the full stack
Languages
Machine learning & data
Platforms & engineering
Pandas · NumPy · Matplotlib · Seaborn · Tableau · Power BI · Streamlit · Neo4j · Jupyter · OCI
| Highlight | |
|---|---|
| 🎓 | M.Sc. Data Science, Vellore Institute of Technology - AP (2024-2026) |
| 🏆 | Qualified TCS NQT aptitude and coding rounds; selected for TCS Digital - AI Cloud |
| ☁️ | Oracle Cloud Infrastructure 2025 Generative AI Professional |
| 📊 | AWS Cloud Essentials and Accenture Data Analytics & Visualization simulation |
| 🧠 | Advanced Machine Learning certification from NIT Warangal |
I'm open to Data Science, Machine Learning, and Analytics opportunities, as well as research and open-source collaborations.
Designed with curiosity, evidence, and a bias toward building.