Rohit J.

Key Skills

Software

Other

Top Subject Matter

Machine Learning Engineer

Software Engineer (Fullstack)

Natural Language Processing

Top Data Types

Text

Image

Computer Code Programming

Top Task Types

Classification

Object Detection

Freelancer Overview

Sarcasm Detection Data Annotation and Text Classification. Brings 3+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal, Proprietary Tooling, and Other. Education includes Bachelor of Technology, Kalinga Institute of Industrial Technology (2025). AI-training focus includes data types such as Text and Image and labeling workflows including Classification and Object Detection.

Labeling Experience

License Plate Recognition Data Annotation and Object Detection

OtherImageObject Detection

Developed a machine learning-based license plate recognition system by combining object detection and OCR technologies. Annotated and labeled large volumes of vehicle image datasets with bounding boxes for license plates and text for OCR tasks. Deployed models and validated ground truth for both object detection and character recognition, iteratively improving label quality. • Amalgamated pre-trained and custom models for robust detection labeling. • Used transfer learning (VGG-16) for precise bounding box annotation. • Leveraged Firebase to log model outputs and annotation revisions. • Focused on real-world accuracy of annotated license plate images.

Not specified

Object Detection Labeling for BNNs (CIFAR-10 and MNIST)

OtherImageObject Detection

Developed binary neural networks and applied them to standard vision datasets such as CIFAR-10 and MNIST for object detection and classification labeling tasks. Created, labeled, and annotated image datasets for training and evaluating the performance of binary neural network models. Validated improvement in accuracy rates by segmenting and classifying objects within the images. • Used Larq library and Python environment to build and test models on labeled datasets. • Performed manual and automated labeling to prepare ground truth data. • Segmented images and annotated objects to train supervised models effectively. • Achieved high accuracy and reduced processing time by refining labeling processes.

Not specified

Sarcasm Detection Data Annotation and Text Classification

TextClassification

Worked on detecting sarcasm in text using natural language processing techniques, focusing on tokenization, word embeddings, and word vectorization. Developed machine learning models to classify movie description text entries as sarcastic or not. Created a labeled dataset based on scraping and preprocessing text data from IMDb sources. • Applied Exploratory Data Analysis to clean and annotate over 30,000 text samples. • Implemented NLTK and custom labeling pipelines for entity classification. • Achieved 90% labeling precision through iterative training and evaluation. • Leveraged an in-house labeling setup with Python and Pandas.

Not specified

Education

Work History