Mahad M.

Key Skills

Software

Other

Don't disclose

OpenCV AI Kit (OAK)

Top Subject Matter

Autonomous driving / computer vision

Geospatial computer vision / roadway assets

AI tutoring / educational content generation

Top Data Types

Video

Document

Image

Top Task Types

Object Detection

Geocoding

Text Generation

Freelancer Overview

Senior Machine Learning Engineer. Brings 7+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include TensorFlow, Other, and Don't disclose. Education includes Bachelor of Science, N/A (2024). AI-training focus includes data types such as Video, Geospatial, and Tiled Imagery and labeling workflows including Object Detection, Geocoding, and Text Generation.

Labeling Experience

Identifying Geolocation of Roadway Assets from 2D Images (Mar 2024 – Present)

OpenCV AI Kit (OAK)VideoGeocoding

Created an AI-driven computer vision pipeline to identify geolocation of roadway assets from 2D images/video footage. Developed a YOLO-based object detection approach for road sign classification from continuous video streams. Produced geospatially accurate road sign coordinates and integrated them with downstream GIS systems. • Detected and classified road signs using YOLO • Fine-tuned depth accuracy for roadway imagery • Generated geospatial coordinates for roadway assets • Built production-grade end-to-end processing pipelines

2024 - Present

Senior Machine Learning Engineer, Paklogics (May 2025 – Present)

VideoObject Detection

Worked on object localization and classification for self-driving car applications using deep learning models in TensorFlow. Adapted state-of-the-art computer vision algorithms to improve pedestrian, vehicle, and traffic sign recognition from visual inputs. Contributed to model-ready datasets and evaluation workflows to support detection accuracy improvements. • Localized objects relevant to autonomous driving scenarios • Classified pedestrians, cars, and traffic signs • Adapted deep learning architectures for improved recognition • Supported end-to-end development from data to evaluation

2024 - Present

You Learn – AI Powered Learning Tool (Jan 2025)

Don't discloseDocumentText Generation

Developed an AI-powered learning tool that generates adaptive flashcards and quizzes from user-provided content such as YouTube videos, PDFs, PPTs, and web links. Used NLP and semantic chunking to convert source material into study-ready prompts and content. Implemented retrieval-augmented generation to produce precise, user-specific learning outputs. • Produced intelligent flashcards from input materials • Generated adaptive quizzes from semantic chunks • Integrated RAG with embeddings and semantic search • Delivered contextual, user-specific outputs

2025 - 2025

Machine Learning Engineer, Webtechsofts (Jan 2024 – March 2025)

OtherGeocoding

Built and deployed a geospatial AI solution to automate geotagging of roadway assets from vehicle-mounted camera footage. Processed continuous visual inputs into geospatially aligned outputs to reduce asset management costs. Supported downstream mapping integration by producing location-ready coordinates and attributes. • Automated geotagging of road signs • Generated geospatial outputs from camera footage • Improved roadway asset management efficiency • Enabled integration with downstream geospatial systems

2024 - 2025

Education

Work History