Machine Learning Engineer | AI/ML Enthusiast | Problem Solver

📍 New Delhi | 📧 vinodadi987@gmail.com | 📞 +91 7836062458
🔗 LinkedIn | 💻 GitHub

AI/ML Engineer with 2+ years of hands-on experience in building solutions across Deep Learning, NLP, GenAI, Computer Vision, and Agentic AI. Skilled in end-to-end model development, data pipelines, and deployment. Passionate about applying AI to solve real-world challenges.

Programming: Python, C++, C, Flutter, SQL, Dart
AI/ML: CNN, RNN, LSTM, NLP, TensorFlow, PyTorch, LLMs, LangChain, LangGraph, RAG
Deployment: FastAPI, Flask, Streamlit, Firebase, SQL, Docker, Heroku
Libraries & Tools: Mediapipe, Pandas, NumPy, Seaborn, Matplotlib, Scikit-Learn, Hugging Face
Soft Skills: Leadership, Problem Solving, Quick Learning

AI/ML Intern – Medoc Health (Remote) | Present

• Built OCR + Summarization pipeline for doctor’s prescriptions with +5% extraction accuracy.
• Implemented RAG + Extractive Segment Retrieval for pharma data; deployed in Tier-2/3 partner hospitals.

AICTE Virtual Internship – Edunet Foundation | Jan 2024 – Feb 2025

• Built and deployed Tree Classification models using Vision Transformers & Streamlit.

AI/ML Trainee – NIELIT, Ropar | May 2024 – Jun 2024

• Developed ML models with impactful EDA insights.
• Deployed a Flask-based Calories Burnt Prediction App with 94% accuracy.

• Goal: Eliminate need for human interpreters in offline stores.
• Tech: CNN + BiLSTM, Flutter, Firebase, TensorFlow
• Accuracy: 95% real-time sign recognition

• Goal: Movie recommendation using Collaborative + Content-based filtering
• Tech: Flask, Python APIs
• Features: Collaborative Filtering, Content-Based Filtering, Hybrid Models, MovieLens Integration

• Goal: Predict calories burnt during exercise using physiological data
• Tech: Streamlit, Flask, ML Regression, EDA
• Accuracy: 94%

• Goal: Apply causal inference methods to energy datasets
• Tech: Python, Jupyter Notebooks, Do-Calculus, A/B Testing, Propensity Scoring, Causal ML
• Features: Step-by-step notebooks, real-world datasets, end-to-end causal analysis pipeline

• Goal: Predict churn based on customer demographics & usage
• Tech: ANN, Keras, Google Colab | Accuracy: 86%

• Goal: Predict relevant tags for documents using ML models
• Tech: NLP, ML pipelines

• Goal: Built advanced Agentic AI pipelines for contextual reasoning and multi-step task automation.
• Tech: LangChain, LangGraph, OpenAI APIs, RAG, Vector DBs (Pinecone/FAISS).
• Impact: Delivered contextual document understanding and Q&A over enterprise-scale datasets.

• Goal: Automated doctor prescription digitization with OCR & extractive summarization.
• Tech: Tesseract OCR, Hugging Face Transformers, PyTorch, FastAPI.
• Impact: Achieved +5% extraction accuracy over baseline; integrated into hospital workflows.

• Goal: Built a domain-specific Extractive Question Answering system for structured knowledge retrieval.
• Tech: Hugging Face QA Models, XGBoost (for hybrid ranking), RAG pipelines.
• Impact: Improved answer accuracy by +12% on benchmark datasets.

BTech in Computer Science – IKGPTU, Kapurthala
Oct 2022 – May 2026
Relevant Coursework: Machine Learning, Deep Learning, AI, OOPs, DSA, Cloud, OS, DBMS, CN

• 🏆 RedBus Hackathon Finalist
• 🏆 NCC RDC Camp – Selected
• 🤖 Top-5 International Robotics Competition
• 🏆 Kaggle Expert

I love collaborating on open-source, hackathons, and research projects.
If you're a recruiter, fellow developer, or enthusiast, feel free to reach out for collaboration, hiring, or just to say hi!

⭐️ From Aditya Vashist