Mayank C.

Key Skills

Software

Other

Top Subject Matter

Civic technology

road pothole detection

Top Data Types

Image

Top Task Types

Object Detection

Classification

Freelancer Overview

AI & Backend Developer (Parul University – VMC Collaboration). Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Bachelor of Technology, Parul University (2027) and Higher Secondary, St. Francis Hr. Sec. School (2023). AI-training focus includes data types such as Image and labeling workflows including Object Detection and Classification.

Labeling Experience

AI and Backend Developer - Parul University

ImageClassificationObject Detection

Worked as an AI and backend developer for a university-industry civic technology collaboration focused on computer vision and AI-driven civic analytics. Developed a YOLOv8-based pothole detection model and exposed results through a FastAPI REST endpoint. Built a LangChain-based NL-to-SQL system with schema-aware prompting to translate administrator queries into structured database outputs.• Fine-tuned YOLOv8 using a manually collected and annotated pothole dataset and produced bounding-box coordinate predictions. • Implemented FastAPI endpoints for image input and returning annotated outputs. • Designed an NL-to-SQL agent with schema-aware prompting and returned structured civic database results. • Managed end-to-end pipeline work including data collection, model training, API development, and institutional review submission.

2025 - 2026

AI & Backend Developer (Parul University – VMC Collaboration)

OtherImageObject Detection

Fine-tuned a YOLOv8 object detection model using a manually collected pothole image dataset with annotated targets. Built an image-to-bounding-box API that returns coordinates and annotated output for road images. Owned the end-to-end workflow from dataset creation through model training and API integration for civic technology review. • Data preparation with manual annotation for pothole imagery • Training and evaluation of a YOLOv8 detector • FastAPI inference endpoint producing bounding-box coordinates • Submission of the full system as part of a university-industry civic collaboration.

2025 - 2026

Education

Work History