ANKITA NAIK
Hyderabad
Summary
AI Engineer with hands-on experience in computer vision, deep learning, and time-series forecasting, currently working on real-time, production-grade vision systems. Strong background in building end-to-end ML pipelines—from data collection and annotation to model training, optimization, and deployment. Experienced in YOLO-based object detection, Vision Transformers, LSTM forecasting, and automated log analysis. Passionate about applying AI to solve real-world problems, with research exposure in medical imaging, and a published dataset in retinal disease analysis.
Overview
3
3
years of professional experience
1
1
Certification
Work History
Junior AI Engineer
Retearn
04.2025 - Current
- Designed a real-time vision-based anti-pinch detection system to identify object intrusion in a restricted mechanical flap zone.
- Implemented polygon-based ROI segmentation configurable via JSON to support multiple camera orientations and flip modes.
- Developed a hybrid detection pipeline combining contour-area deviation analysis with pixel-wise background differencing.
- Achieved stable high-FPS performance (up to 120 FPS) using temporal trigger logic and ROI-bounded processing on an embedded vision pipeline.
- Built an end-to-end real-time object detection system using YOLOv5.
- Managed dataset collection, annotation, preprocessing, model training, and evaluation.
- Optimized inference speed and detection accuracy for production-ready deployment.
- Developed a deep learning-powered image retrieval system for brand detection and similarity search.
- Used Vision Transformer (ViT-B/16) for robust feature extraction and FAISS (IVF index) for fast, scalable nearest-neighbor search.
- Integrated YOLO-based label region extraction to crop brand logos before feature embedding.
- Built a Flask-based web interface supporting image search, similarity scoring, and incremental indexing.
- Designed and implemented an LSTM-based time-series model to forecast machine transactions.
- Performed predictive maintenance analysis on historical machine usage data.
- Enabled early detection of abnormal patterns and potential machine issues.
- Developed a Python-based automated log analysis and ETL pipeline.
- Parsed large-scale transaction, device, and item-level machine logs.
- Generated structured analytical reports using the Google Sheets API for monitoring and debugging.
- Analyzed image quality degradation and restoration during compression and decompression.
- Applied classical interpolation techniques to study resolution loss.
- Evaluated performance and visual trade-offs across different methods.
- Tools: OpenCV, YOLOv5, Vision Transformer (ViT), FAISS, Flask, LSTM, NumPy, GStreamer, Computer Vision, Time-Series Analysis, Log Analysis, ETL, Pandas, regex.
Data Engineer Intern
Retearn
09.2024 - 04.2025
- Designed and implemented an object detection pipeline using the YOLOv5 deep learning model to accurately identify and localize objects in images and video frames.
- Conducted end-to-end dataset creation, including data collection, annotation, and preprocessing, to ensure a diverse and balanced dataset for training. I analyze logs to optimize performance, troubleshoot issues.
Internship
Central of Excellence, SGGS&IT
07.2023 - 01.2024
- Conducted in-depth research on advanced AI algorithms, focusing on machine learning and deep learning methodologies.
- Actively participated in cutting-edge projects aimed at enhancing diagnostic tools for medical imaging, specifically in retinal disease detection. Worked closely with academic supervisors and advanced the field of AI in medical imaging.
AI Intern
Ziegler Aerospace
11.2023 - 12.2023
- Learned and applied advanced AI technology for object extraction, text extraction. Conducted research on Machine learning and Deep learning methodologies.
Education
M.Tech - Artificial Intelligence
Shri Guru Gobind Singhji Institute of Engineering And Technology (SGGSIET)
Nanded07.2023
B. Tech - Electronics and Telecommunication
Dr. Babasaheb Ambedkar Technological University
Lonere06.2021
Skills
- ML with Python
- TensorFlow
- Keras
- Object Detection
- YOLO
- Pandas
- Opencv
- Numpy
- Scikit-learn
- Matplotlib
- Deep Learning
- Computer vision algorithm development
- LLM
- Time-series analysis
- Image processing
- Data preprocessing
- Log analysis
- TensorFlow framework
- Statistical modeling
- Python programming
- Neural networks
- Model evaluation
- Feature engineering
- PyTorch framework
Certification
Machine Learning with Python, Coursera, 06/01/22
Projects
Retinal blood vessel segmentation, accurately identify and segment blood vessels in retinal images using advanced segmentation techniques. Retinal multi-disease classification using deep learning algorithms, perform image classification to classify retinal diseases
Publications
Retinal Fundus Multi-Disease Image Dataset (RFMiD) 2.0: A Dataset of Frequently and Rarely Identified Diseases., Data, 8, 2, 29, MDPI, https://www.mdpi.com/2098772
Timeline
Junior AI Engineer
Retearn
04.2025 - Current
Data Engineer Intern
Retearn
09.2024 - 04.2025
AI Intern
Ziegler Aerospace
11.2023 - 12.2023
Internship
Central of Excellence, SGGS&IT
07.2023 - 01.2024
M.Tech - Artificial Intelligence
Shri Guru Gobind Singhji Institute of Engineering And Technology (SGGSIET)
B. Tech - Electronics and Telecommunication
Dr. Babasaheb Ambedkar Technological University