Hi, I'm

AI/ML Engineer • Deep Learning • LLMs & Automation • MLOps
Download CV
AI/ML Illustration

About Me

Twisha Patel

I am a dedicated AI/ML Engineer passionate about building impactful solutions using deep learning, computer vision, NLP, and MLOps. I enjoy working on challenging real-world problems—ranging from object detection pipelines to fine-tuning LLMs for business & healthcare domains. I thrive in collaborative teams, believe in lifelong learning, and love turning emerging technologies into practical value.

My Skills

Python JavaScript PHP MySQL
HTML5 CSS3 Bootstrap Git & GitHub
TensorFlow PyTorch OpenCV Hugging Face
NLP Computer Vision FastAPI Docker CI/CD
LLMs Prompt Engineering Model Deployment

Education

Advanced Post Graduate
Agentic AI Workflows & Agentic System Development
Indian Institute of Technology, Madras
Dec 2025 – Jun 2026
Ongoing
Bachelor's Of Technology
Computer Science Engineering
ITM (SLS) Baroda University
Aug 2021 – May 2025
CGPA: 8.6
Higher Secondary School
Class XII
Shri J.R. Shah Bright Eng. Medium School
2020 – 2021
Grade: B2
Secondary School
Class X
Bright Day School
2019 – 2020
Grade: B1

Experience

Associate AI/ML Engineer @ Sundaram Technologies
Jul 2025 – Present Vadodara
  • Working on object detection and computer vision projects for industrial automation.
  • Developing Python-based ML and deep learning solutions for image/video analysis.
  • Responsible for data analysis, preprocessing, and model evaluation to enhance detection accuracy.
  • Leveraging frameworks such as TensorFlow, OpenCV, and PyTorch for real-world vision applications.
AI/ML Intern @ Artem HealthTech
Jan 2025 – Jul 2025 Ahmedabad
  • Built machine data integration solutions for the healthcare platform (automation & data flow).
  • Deployed real-time object detection using MMDetection's pre-trained model on live webcam feeds.
  • Developed agentic speech-to-text pipelines using open source models.
  • Fine-tuned finance-specific LLM (Flan-T5), optimized prompts for domain accuracy.

Python & AI/ML Projects

Real-world AI, ML & Automation: What I've Built

Open Source Contribution – LlamaIndex

Open Source 2026

  • Contributed a FastAPI-based Retrieval-Augmented Generation (RAG) example using LlamaIndex with fully local Ollama LLMs.
  • Implemented document ingestion, vector indexing, and query APIs enabling offline, no-API-key RAG workflows.
  • Collaborated with maintainers to address architecture, formatting, and CI feedback; contribution merged into main repository.
Tech: Python, LlamaIndex, FastAPI, Ollama, Local LLMs, RAG
View Repository

Pneumonia Detector AI

Medical Imaging 2025

  • Developed an explainable medical AI web app for automated pneumonia detection on chest X-rays, leveraging EfficientNet and GradCAM for visual model transparency and user trust.
  • Implemented a human-in-the-loop feedback and retraining pipeline with batch processing, quality assurance, and full Dockerization for reproducible deployment in clinical or research environments.
Tech: Python, PyTorch, EfficientNet, Streamlit, GradCAM, Docker
View Code

AutoML Pipeline Service

AI Automation 2025

  • Designed an end-to-end pipeline using Streamlit & FastAPI for dynamic model training, inference, and benchmarking.
  • Automated dataset profiling, model selection, versioning, and performance logging.
  • Web interface for configuration, easy deployment & result visualization.
Tech: Python, Streamlit, FastAPI, scikit-learn, Docker
View Code

Object Detection

Computer Vision 2025

  • Implemented real-time object detection with MMDetection & OpenCV on live webcam feeds.
  • Displayed bounding boxes & class labels with pretrained COCO weights.
  • Hands-on with MMDetection API, model inference & live result rendering.
Tech: Python, MMDetection, OpenCV
View Code

Traffic Sign Recognition

Deep Learning 2023

  • Developed a CNN model to classify German traffic signs with high accuracy.
  • Advanced preprocessing & augmentation with OpenCV for model robustness.
  • Built with TensorFlow & Keras, leveraging GTSRB dataset.
Tech: Python, TensorFlow, OpenCV, Keras
View Code

Finance-based Small LLM

NLP/LLMs 2025

  • Fine-tuned Flan-T5 LLM on curated financial Q&A datasets.
  • Optimized prompt engineering for precision & contextual relevance.
  • Deployed model for domain-specific finance Q&A.
Tech: Python, Hugging Face, Flan-T5, Transformers
View Code

Voice Assistant

Speech AI 2023

  • Developed a desktop voice assistant in Python for reminders, music, web search & more.
  • Integrated speech recognition and TTS for seamless two-way interaction.
  • Enhanced natural language understanding for smarter commands.
Tech: Python, SpeechRecognition, pyttsx3
View Code

Contact

Interested in working together or just want to connect? Reach out anytime.

Phone
+91 8160893358
Email
twishap534@gmail.com
Location
Ahmedabad, Gujarat, India