Mohammed M.

Key Skills

Software

Other

Label Studio

SuperAnnotate

Telus

Top Subject Matter

Customer review analytics

NLP sentiment/emotion

topic modeling

Top Data Types

Text

Image

Document

Top Task Types

Emotion Recognition

Object Detection

Data Collection

Text Generation

Text Summarization

Transcription

Freelancer Overview

Customer Review Sentiment Analysis & Topic Modeling (GenAI) — NLP pipeline and modeled customer emotions/defects. Core strengths include Hugging Face, PostgreSQL, and PyTorch. Education includes Master of Science, University of Sussex (2024) and Bachelor of Science, MS Ramaiah University (2023). AI-training focus includes data types such as Text, Computer Code, and Programming and labeling workflows including Emotion Recognition, Evaluation, and Rating.

Labeling Experience

Retail Shelf Monitoring Using Convolutional Neural Networks — trained detection model

ImageObject Detection

Trained a computer vision model to detect out-of-stock items and misplaced products on retail shelves. The training required creating/using labeled visual examples and evaluating detection performance for reliable deployment. Completed the end-to-end project lifecycle with a focus on model-driven detection accuracy and clear documentation. • Trained ResNet-50 CNN for retail shelf item state detection. • Validated and iterated on detection results to reduce manual inspection. • Managed data collection, model training, evaluation, and documentation. • Ensured outputs were suitable for scalable, proactive monitoring.

2024 - 2024

Advanced Retail & Supply Chain Analytics — SQL/Python segmentation and reporting

Engineered analytical pipelines to profile purchasing behavior and produce structured customer segments for downstream decision-making. The project focused on transforming raw data into labeled customer groups and operational reporting outputs. Produced written and visual summaries that mapped model-driven segmentation to stakeholder actions. • Performed customer behavior profiling using advanced SQL techniques. • Applied RFM segmentation logic to derive categorical customer labels. • Automated reporting workflows to reduce manual analysis effort. • Generated documentation and stakeholder-ready summaries.

2024 - 2024

Customer Review Sentiment Analysis & Topic Modeling (GenAI) — NLP pipeline and modeled customer emotions/defects

TextEmotion Recognition

Built an NLP pipeline to process thousands of unstructured customer reviews for automated sentiment and emotion-related outputs. The work included preparing and iterating on text inputs and validating model outputs against recurring defect signals. Designed the overall dataset-to-insight flow so results could be consistently reproduced for analysis and stakeholder use. • Ingested large volumes of customer review text from an e-commerce source. • Implemented sentiment/emotion detection using transformer-based models. • Identified recurring themes and product-defect patterns from text. • Documented data flow and output interpretation for reproducibility.

2024 - 2024

Large-Scale Astronomical Data Clustering — MSc Thesis (Unsupervised ML)

OtherData Collection

Independently designed and executed an unsupervised ML pipeline to cluster high-dimensional unlabelled spectral data on HPC systems. The project involved organizing spectral datasets and producing cluster assignments from model outputs rather than manual annotation. Work included handling technical failures, adapting workflows, and delivering results to strict academic deadlines. • Prepared and ran HPC-based clustering workflows on high-dimensional spectral data. • Generated cluster assignments from unlabelled scientific inputs. • Adapted pipelines in response to technical issues during execution. • Produced structured progress updates and final outcomes for supervisors.

2023 - 2024

Education

Work History