Pavan N

AI and Machine Learning professional with extensive experience in automating data-intensive processes and delivering AI-driven solutions across diverse sectors, including healthcare, FMCG, oil & gas, and engineering. Adept at leveraging deep learning models, OCR, and document intelligence techniques to drive efficiency and reduce manual effort. Proven track record in successfully implementing solutions that enhance operational workflows and ensure high accuracy in data extraction and analysis.

1. AI Powered Ordering Assistant.

· Built an AI-powered WhatsApp ordering assistant that automated order placement, summarization, and confirmation for B2B customers.

· Developed a reorder reminder scheduler that analyzed order history, generated personalized WhatsApp eminders, and triggered reorder flows.

· Improved data reliability by normalizing CSV -> JSON ingestion, added unit tests and integrated socket.io for real-time tracking, reducing manual follow-ups and improving customer engagement.

· Tools: Python, Flask, OpenAI, SQLite, Langchain, Prompt optimization.

2. Comparison of Digital Packaging Labels (FMCG Sector)

· Developed a computer vision and NLP-based system for automated comparison of digital packaging artwork using named entity recognition (NER) for text and symbol matching.

· Integrated rule-based matching and similarity scoring for compliance checks.

· Reduced manual quality assurance effort by over 95%, accelerating go-to-market time for product labels.

· Tools: Python, OpenCV, spaCy, Pandas, Scikit-learn, Object detection, Image classification and segmentation.

3. Digitization of Piping and Instrumentation Diagrams (Oil & Gas)

· Automated extraction of Regions of Interest (ROIs), inlet/outlet classification, and custom object detection (valves, pumps, arrows) from engineering drawings.

· Designed deep learning models for arrow orientation and direction detection using annotated datasets.

· Enabled intelligent CAD model generation and tagging, reducing manual digitization effort by ~90%.

· Tools: PyTorch, OpenCV, YOLO, OCR, custom annotation tools, object detection, image classification.

4. Engineering Bill of Materials (EBOM) Analysis

· Applied OCR (Tesseract, EasyOCR) and NLP rule-based engines to extract structured metadata from scanned engineering documents.

· Integrated GPT-based models to interpret and reformat BOM tables, retaining source formatting across PDFs, Excel sheets, and images.

· Improved document conversion accuracy and consistency by ~30%, reducing post-processing workload.

· Tools: Python, GPT APIs, LangChain, Pandas, Tesseract, PaddleOCR.

5. Spare Parts Recommendation System

· Designed and deployed a transformer-based recommendation model for predicting next-item purchases using historical customer behavior data.

· Implemented a full ML pipeline for data ingestion, preprocessing, training, hyperparameter tuning, and inference.

· Improved top-N recommendation precision by ~20%, aiding supply chain teams with actionable insights.

· Tools: PyTorch, Hugging Face Transformers, MLflow, FastAPI, Feature Engineering.

6. Intelligent Document Processing & Retrieval System

· Built an end-to-end document understanding platform combining OCR, RAG (Retrieval-Augmented Generation), and LLM-based pipelines.

· Features included semantic search, summarization, key-value extraction, and contextual Q&A from large sets of PDFs.

· Designed a Streamlit-based UI for batch processing, result visualization, and logging.

· Delivered a scalable, automated document analysis tool reducing manual review time by ~85%.

· Tools: LangChain, OpenAI GPT, FAISS, Streamlit, PyMuPDF, Pandas, Transformers, customer segmentation, embedding generation.

• Certificate of Recognition for filing a patent at Indian patent Office titled "Method & System of Comparing Digital Labels" (IN202141039416), awarded by L&T Technology Services.
• "Method and System for generating editable schematic illustration of non-editable schematic illustration" (IN202541019135)
• "Method and System for determining pipeline flow in non-editable schematic illustration" (IN202541025242)
• "Method and System for correcting skewness in images" (IN20254103399)

Microsoft Certified, Azure AI Engineer - Microsoft.

• Strong desire to stay updated with the latest advancements in AI/ML technology.
• Actively contributes to team culture fostering diversity, equity, and inclusion.