Summary

AI Engineer / Senior Software Engineer with 7+ years of experience in machine learning, computer vision, and NLP.

Skills:

• Programming Languages: Python, JavaScript, Java, C++
• Web Frameworks & Servers: FastAPI, Flask, AngularJS, Angular 2/4, NodeJS, REST API Integration, Gunicorn, Uvicorn, Nginx
• Libraries & Tools: TensorFlow, TensorFlow Lite, Keras, PyTorch, OpenCV, NumPy, Matplotlib, Scikit-Learn, NLTK, SpaCy, MMDetection, ZenML
• Machine Learning & NLP: Named Entity Recognition (NER), Text Classification, Pose Estimation, Semantic Search, Recommendation Systems, Hybrid Filtering, PDF Parsing, RAG LLM Applications, AI Text Detection, Passive–Active Sentence Conversions
• Computer Vision: Object Detection, Image Preprocessing, Compound Image Analysis, Facial Landmark Detection, License Plate Detection, MobileNet SSD, Faster R-CNN, RetinaNet, Cascade R-CNN
• Mobile, Desktop & Edge Development: Android, TensorFlow Lite, Ionic, React Native, Tkinter (Desktop UI)
• IoT & Communication Protocols: MQTT, TCP, Real-Time Sensor Data Processing
• Databases: SQLite / SQL Databases
• DevOps & ML Deployment: CI/CD Pipelines, AzureML, Production-Grade Server Setup, Cloud Deployment (Azure, AWS, Firebase)
• Data Processing: EDA, Annotation Tools (LabelImg, VoTT, SuperAnnotate), Feature Engineering
• Core Strengths: Strong Python Programming, Debugging & Problem-Solving Skills

Work Experience:

Machine Learning Engineer, Ahmedabad, India
July 2021 - Present

• Developed Semantic Search Engines for scientific document retrieval.
• Designed and implemented Journal Recommendation Systems for publishing companies.
• Created custom Named Entity Recognition (NER) and Span Categorization solutions for scientific documents.
• Built NER Visualization and Tagging Tools to facilitate user-friendly annotation and review.
• Delivered advanced Computer Vision solutions for Compound Image Analysis, enabling image splitting and processing in scientific documents and research papers.
• Managed DevOps workflows, including scalable model training, testing, and deployment on high-performance servers.
• Engineered modules for extracting and processing abbreviations, acronyms, citations, paraphrases, and references from PDFs and images.
• Led development of passive-to-active and active-to-passive sentence conversion tools for scientific and research texts using LLMs, including deployment, result evaluation, and rule-based validation.
• Designed and implemented masking solutions for sensitive information in patent research.
• Developed AI-powered Patent Processing Modules.∙Built Retrieval Augmented Generation (RAG) LLM Applications tailored for patents, journals, and research organizations.
• Developed robust PDF parsing solutions including figure, caption, table, layout extraction, and reference mapping with in-depth analysis.
• Spearheaded Patent Infringement Detection, Localization, and comprehensive analysis systems, achieving top-tier industry standards.
• Implemented AI-Generated Text Detection solutions with performance comparable to leading platforms like CopyLeaks.

ML Engineer and Team Leader

Ahmedabad, India | Dec 2019 - Jul 2021

• Developed a Digital Document Process Automation solution (PDF 3.0) for high-accuracy data and entity extraction from multilingual, multi-format PDFs.
• Led data collection, exploratory data analysis (EDA), validation, preprocessing (including computer vision techniques), and feature engineering.
• Managed large-scale annotation efforts using tools like LabelImg, VoTT, and SuperAnnotate for over 30,000 PDF images.
• Applied MMDetection (PyTorch framework) for object detection tasks including entities, key-value pairs, and tables.
• Implemented and tested multiple deep learning models including FasterRCNN, RetinaNet, and CascadedRCNN.
• Applied NLP techniques for text classification and NER using SpaCy.
• Set up production-grade web servers using Nginx, Gunicorn, and Uvicorn, with FastAPI and Flask for robust backend services.
• Led MLDevOps pipeline configurations for automated training and deployment (CI, CT, CD) and deployed models on AzureML.
• Conducted training sessions for junior engineers and participated in AI recruitment.
• Managed system integration, frontend and API testing to ensure production readiness.

Software Developer (Machine Learning),
April 2018 – Nov 2019

• Built computer vision–based exercise classification and repetition counting software using pose estimation, with strong Python programming for core logic and data processing.
• Developed real-time vehicle license plate detection and recognition using smartphone cameras, covering localization, image preprocessing, OCR, and symbol recognition, with REST API integration for backend services.
• Created a Snapchat-like mobile application featuring facial landmark detection using Google Mobile Vision APIs.
• Implemented exercise classification using pose estimation for both desktop and mobile platforms, including a Tkinter-based desktop UI.
• Designed and optimized mobile object detection and localization models using MobileNet SSD for real-time performance.
• Developed human activity classification systems using real-time accelerometer and gyroscope data, leveraging MQTT & TCP for IoT communication.
• Built Android applications for real-time human body pose detection using pose estimation frameworks and Python-based preprocessing pipelines.
• Implemented MobileNet V2–CPM models using TensorFlow APIs, deployed on Android devices via TensorFlow Lite, and optimized for low-latency inference.
• Created a real-time body pose correction and audio guidance system, with sensor data persistence using SQLite / SQL databases.

Jr. JavaScript Developer (Full Stack), Ahmedabad, India
May 2017 – Nov 2017

• Developed web and mobile applications using JavaScript, Angular 4, NodeJS, Ionic, and Firebase for international clients.

Certification & Courses:

• Deep Learning Specializationby Andrew Ng (Coursera, Deeplearning.ai)
  Feb2018 - Jul 2019
• Machine Learning by Andrew Ng (Stanford Online)
  Sep 2018 - Jan 2019

Project Highlights:

• 3D Printer cum Robotic Arm| Dec 2017 - Jul 2018
• Designed 3D slicing, visualizing, and rendering software. Built a robotic arm using Autodesk Fusion 360 and RepRap open-source hardware.
• Smart House Automation (IoT) | Dec 2016 - Apr 2017
• Integrated Arduino, NodeMCU-ESP8266, and Raspberry Pi 2 for home automation with touch and voice commands. Provided real-time data via a mobile application.
• 3D Rubik’s Cube Game Application | Apr 2016 - Nov 2016
• Developed an Android-based 3D Rubik’s Cube simulation and game using JavaFX.
• IR Robot (Line Following Robot)| Feb 2015
• Created a line-following robot capable of tracking black lines on white backgrounds under various lighting conditions.