Share Profile
Zaid Bin Khalid
Aspiration Statement
I am pursuing roles at the intersection of product and technology, ideally at startups where I can grow from a technical contributor to someone shaping what gets built and why.
Core Skills
- Agentic AI & Automation
- Artificial Intelligence
- Data Science
- Machine Learning & Deep Learning
- Python
- Robotics & Embedded Systems
Core Competencies
- Adaptability
- Collaborates Openly
- Drive for Results
- Effective Presentation Skills
- Encourages Innovation
- Planning
- Problem Solving
- Takes Initiative
Preferred Career Paths
First priority: Product Manager
Second priority: AI Engineer
Third priority: Software Engineer, Data Analyst, Data Engineer
Experience
Leadership / Meta-curricular
- Event Lead, Ieee (Institute of Electrical and Electronics Engineers)
- Treasurer, Math Club
Internship / Volunteer Work
- Teaching Assistant For Calculus I, Habib University (September – December 2024)
- Summer Tehqeeq Student Researcher, Habib University (June – August 2024)
- Peer Tutor For Calculus Ii, Habib University (February – April 2024)
- Teaching Assistant For Calculus I, Habib University (August – December 2023)
Publications / Creative Projects
- Conference Presentation – Zaid Bin Khalid, Nahyan Javed, Farhan Khan. "BiLSTM-Based Sequential Acoustic Crash Detection for Urban Road Safety". IEEE Karachi Humanitarian Technology Conference (KHI-HTC), 2026. (Conference Paper Under Review)
- Publication – "Acoustic Car Crash Detection Using Audio Classification". DURS, Habib University, 2026. (Abstract Under Review)
Final Year Project
Project Title
Real-Time Car Crash Detection Using Acoustic Signatures
Description
Road traffic accidents claim 1.19 million lives annually, with delayed emergency response being a critical contributing factor. This project develops a passive acoustic vehicle crash detection and localization system intended for deployment on embedded hardware in low-resource urban environments. In the first phase, three audio classification approaches were evaluated — a Bidirectional LSTM, an SVM with Bag of Aural Words, and a Gaussian Naive Bayes classifier — with the BiLSTM achieving 97.14% accuracy and 98.5% crash recall. The ongoing second phase focuses on two objectives: implementing the detection pipeline on a Xilinx Zynq-7000 FPGA for real-time embedded inference and developing an acoustic localization algorithm to estimate crash coordinates for first responders, completing a fully deployable crash notification system.
Project Pictures
No project pictures available.