Aspiring ML / AI engineer — B.Tech (CSE, Specialization: AI & ML) at Bennett University. I build practical deep-learning solutions and deploy them as user-friendly Streamlit apps. Passionate about medical imaging, generative models, and production-ready ML tooling.

• 📫 Reach me: [email protected] | +91 83189 83400
• 🔗 Links: GitHub • LinkedIn • LeetCode
  (Replace the placeholders above with your actual profile URLs)

I enjoy turning research and prototypes into usable apps. My recent work focuses on medical imaging, GANs for creative image synthesis, and retrieval-augmented generation for domain-specific chatbots. I prefer Python-based stacks and deploying lightweight web UIs with Streamlit.

Bennett University — B.Tech in Computer Science & Engineering (AI & ML)
September 2022 – May 2026

• GPA: 8.39 / 10

Tools: Python, Streamlit, TensorFlow, NumPy, OpenCV
Feb 2025 – Mar 2025 · Project Link

• Assembled a dual-model pipeline (EfficientNetB3 + U-Net) achieving 95% classification accuracy on 7k+ MRIs.
• Implemented reliable tumor segmentation and a Streamlit app that generates clinical PDF reports — reduced manual reporting time by ~40%.

Tools: Python, Streamlit, TensorFlow, NumPy, PIL
Sep 2024 – Oct 2024 · Project Link

• Built two GAN-based models: one for converting hand-drawn sketches into realistic color images (Pokémon) and another for artistic-style transfers (cubism, impressionism).
• Deployed a real-time Streamlit app handling 1k+ user uploads with <2s average transformation latency.

Tools: Python, Streamlit, LangChain, Gemini API
Jan 2025 – Feb 2025 · Project Link

• Built a semantic-search chatbot over Ayurvedic remedies with ~90% retrieval accuracy.
• Launched a Streamlit interface — improved query-to-response performance by ~30% over baselines (tested on 500+ queries).

Tools: Python, Streamlit, TensorFlow, NumPy, OpenCV
Aug 2025 – Sep 2025 · Project Link

• Developed a CNN-based classifier for ECG images to detect six cardiac conditions (incl. MI, LBBB, RBBB) with ~92% validation accuracy.
• Streamlit app enabled rapid screening of 1k+ ECGs and reduced diagnostic turnaround by ~35%.

Integrating Deep Learning Concepts with Blood Diagnosis — GL Bajaj IEEE Research Conference (Jan 2025 – May 2025)

• Proposed a diagnostic framework using Xception + XceptionResNet trained on 75×75×3 tensors; achieved 92.37% ensemble accuracy for diseases including anemia, diabetes, and thrombosis.
• Reported ~5% improvement over existing methods; currently under IEEE conference review.

• Languages & Databases: Python, SQL, MySQL, MongoDB, JavaScript, HTML, CSS
• Frameworks & Libraries: NumPy, Pandas, Scikit-Learn, Matplotlib, TensorFlow, Keras, OpenCV, GANs, PIL, LangChain, HuggingFace Transformers, OpenAI API
• Tools & Platforms: Git, CI/CD, Streamlit, Jupyter Notebook, PyCharm, VS Code, n8n Automation
• Soft Skills: Problem-solving, effective communication, teamwork, creative thinking, technical writing & presentation

• Deep Learning A-Z: Neural Networks, AI & ChatGPT Prize — Udemy (2024)
• IBM Data Analysis for Machine Learning — Coursera (2024)
• Improving Deep Neural Networks: Hyperparameter Tuning, Regularization & Optimization — DeepLearning.AI (2024)
• Natural Language Processing with Classification and Vector Spaces — DeepLearning.AI (2024)
• Convolutional Neural Networks in TensorFlow — DeepLearning.AI (2024)

• Polishing Tumor X for larger clinical validation and a cloud deployment.
• Expanding Canvas AI to support user-style customization and higher-resolution outputs.
• Contributing more notebooks and end-to-end reproducible pipelines to my GitHub.

• tumor-x — MRI classification & segmentation with a Streamlit reporting UI.
• canvas-ai — GAN-based sketch-to-image and style-transfer demos.
• ayurvedic-chatbot — RAG chatbot for Ayurvedic remedies.
• cardio-care — ECG classification and screening Streamlit app.

If you want to collaborate, see a demo, or just say hi — email me at [email protected] or connect on LinkedIn.