Skip to content
View himanshu9699's full-sized avatar

Block or report himanshu9699

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
himanshu9699/README.md

πŸš€ ML Engineer | Machine Learning | Generative AI | LLMs


🌟 About Me

πŸŽ“ Education: B.E. in Computer Science with Artificial Intelligence – Chitkara University (2021–2025), CGPA 8.7/10

πŸ’Ό Current Role: AI Engineer at Prescience Decision Solutions, Bengaluru

🧠 Passionate about building scalable AI systems, LLM applications, and ML pipelines that solve real-world problems

πŸ† Awards & Recognition: Spot Award, L&D Award, IBM Proficiency Badge (watsonx.data), Advanced CV Certification (DeepLearning.AI)

πŸ“Œ Patents:

Table with Utensil Cleaner – Granted

Height & Blade Adjusting Fan – Filed

πŸ› οΈ Skills

Languages: Python, SQL, JavaScript, C++, Bash ML/DL: Scikit-Learn, TensorFlow, PyTorch, XGBoost, LightGBM, HuggingFace LLMs: OpenAI, LLaMA 2, LangChain, Prompt Engineering, Promptflow, AutoML, Vertex AI Cloud: GCP (Vertex AI, BigQuery), AWS (EC2, S3, Glue), Azure (ADF, Blob, Functions) Data & Big Data: ETL/ELT, Apache Beam, Spark, Hive, Hadoop DevOps & Tools: Docker, GitHub Actions, PostgreSQL, MongoDB, MySQL, Django, Linux

πŸ’‘ Featured Projects

πŸ“„ Production-grade Document Parser Achieved 95% extraction accuracy using Azure Document Intelligence + NLP, reducing manual processing by 85%.

πŸ“˜ Enterprise Summarization Pipeline Built with GPT-4/Claude-3 + RAG, processing 1000+ page reports in minutes instead of hours.

πŸ€– AI Tutoring Chatbot Powered by Socratic algorithms + context management, served 10K+ students with 40% improved learning outcomes.

🍲 Nutri++ – Food Calorie Estimation ResNet-50 CNN trained on Food-101 dataset, achieved 87% accuracy; deployed real-time via Flask + OpenCV.

πŸ’¬ Enterprise Support Chatbot (LLaMA 2) Fine-tuned LLaMA 2-7B with LoRA, deployed with FastAPI + Docker + Redis caching, reducing training time by 60%.

πŸ“« Connect with Me

🌐 Portfolio: https://agrawalhimanshu.com

πŸ’Ό LinkedIn: https://linkedin.com/in/himanshu9699

πŸ’» GitHub: https://github.com/himanshu9699

πŸ“§ Email: [email protected]

πŸ† LeetCode: https://leetcode.com/u/himanshu895522

Pinned Loading

  1. movie movie Public

    Jupyter Notebook

  2. Face-Recognition-System Face-Recognition-System Public

    Python

  3. Himanshu_portfolio Himanshu_portfolio Public

    My portfolio

    JavaScript

  4. Nutri Nutri Public

    Jupyter Notebook

  5. Hostel-Management-System Hostel-Management-System Public

    SCSS