Proficient in building scalable Python REST API frameworks using Django, including development of web applications with integrated APIs for data handling, model training, and inferencing. Experienced in implementing Machine Learning pipelines and deploying Generative AI solutions using LangChain, RAG (Retrieval-Augmented Generation), and MCP (Multi-Component Pipeline) architectures for intelligent document processing and contextual response generation.
Learning Platform to help to Monitoring Learner Activities.
Responsibilities: The Learning Hub is an innovative learning platform designed to enhance the monitoring of learners' activities. Our primary focus is to create an engaging and efficient environment for both learners and educators by incorporating advanced technologies such as machine learning (ML) and web application frameworks like Django. The platform's stand out feature is its facial verification system that ensures the presence and attention of learners during sessions, thereby promoting accountability and participation.
Designed and implemented a Data Lakehouse architecture on AWS, utilising key services such as Redshift, Glue, EMR, S3, DMS, Spark, EC2, Lambda, EventBridge, SNS, KMS, and QuickSight for seamless data ingestion, processing, security, and visualisation.
Experienced in machine learning and Python.
Built a machine learning model-based prediction system to predict the sales of different product SKUs in various regions of the above FMCG business.
Built a credit risk prediction system based on historical loan data and machine learning models to manage the credit risk of banking customers post loan disbursement.
Use of Python libraries and fuzzy logic to perform the deduplication of fuzzy duplicates of a stock trading company in the USA. Identity Detection System based on rules, profiling, and other criteria.
Project Description: Developed AI video analytics for Indian manufacturing industries, enhancing safety, and reducing insurance penalties.
Roles and Responsibilities: Developed AI video analytics solution for leading cement and chemical manufacturers in India. Addressed safety compliance challenges, reducing insurance penalties, and compensation claims. Designed an AI system to detect non-safety measures, such as the absence of safety gear, and fall detection. Conducted thorough testing of the system at client facilities. Deployed a video analytics solution using edge devices like NVIDIA Jetson, implemented in Python. Achieved significant impact, improving safety standards and worker behaviour. Resulted in reduced insurance penalties and worker compensation, due to enhanced safety.
This backend system enables intelligent extraction from legal PDFs using Python, Flask, and Generative AI. It integrates OCR (Pytesseract) for reading documents, LangChain with Google Generative AI for extracting structured data, and vector databases (ChromaDB, FAISS) for semantic search. Anomaly detection ensures accurate outputs, while document comparison highlights content discrepancies. RESTful APIs support integration with any front end. Designed for scalable, real-time processing, the system automates complex data workflows in legal document review, and extraction tasks. Easy setup and modular architecture make it adaptable for enterprise use cases.