Daniya
Sohail
AI / ML Engineer
& Full Stack Developer
I build applied ML systems, model evaluation pipelines, and turn experiments into tools people can actually use. MERN stack at my core — intelligence embedded by design.
About
Developer at the intersection of
full-stack web and practical AI.
With the MERN stack as my core toolkit, I build responsive apps that not only look good but actually solve real user problems. I shape well-structured APIs with Node.js and Express, and craft React interfaces that feel natural to use.
On the AI side, I work with TensorFlow and PyTorch to embed intelligence into products — NLP-driven features and data-backed predictions that make systems feel smarter, not just flashier. Right now I'm diving into agentic workflows with n8n and exploring how low-code automation and AI can work together.
Currently in my 7th semester at NED University of Engineering and Technology, Karachi. CGPA: 3.861.
Experience
MazikGlobal Pakistan
Core Developer
- Built a Customer Ticketing System using Microsoft Power Platform, improving ticket resolution times.
- Integrated Canvas App with unified customer view: login, AI chatbot, feedback, notifications, FAQs, and ticket management.
- Developed a Model Driven App for admins: ticket resolutions, urgency insights, SLA configuration.
- Implemented Power Automate flows for auto-assignment, routing, deduplication, and deadline handling.
- Created interactive Power BI dashboards to visualize ticket metrics and enable data-driven decisions.
Neurocomputation Lab — NEDUET
MERN Stack Intern
- Improved front-end and back-end of Instrux, an energy analytics platform for household electricity monitoring.
- Built dynamic data visualizations for energy usage trends, helping users identify cost-saving opportunities.
- Strengthened authentication via session token management to prevent unauthorized access.
- Optimized data handling logic, reducing load times and improving graph rendering for large datasets.
OctaNet Software Services
Python Intern
- Built a desktop ATM interface using PyQt5 with secure login via user ID and PIN.
- Added core banking features: balance check, withdrawals, deposits, and transfers.
- Implemented transaction history view to help users track activity.
Projects
SolarOps
Solar power management platform optimizing grid placement using 21+ environmental parameters including solar elevation and weather data.
EventLedger
Decentralized event ticketing platform built on Ethereum with NFT-based tickets, smart contracts via Remix, and IPFS validation.
Virtual Radiologist
AI-driven disease detection for pneumonia and tuberculosis using VGG16 and CNN models trained on large chest X-ray datasets.
Complaint Tracking Platform
Web app for managing citizen complaints with role-based dashboards for citizens, administrators, and officers.
Stock Price Prediction
Predictive model forecasting stock prices on a 162k+ record dataset (2001-2022) using PCA dimensionality reduction.
Online Class Marker System
Automated attendance and analytics system for virtual classrooms featuring real-time participation tracking and instructor dashboards.
Data Center Thermal Stability
Thermal distribution simulation for server racks providing predictive cooling recommendations based on airflow and load parameters.
Skills & Recognition
Awards & Scholarships
-
Winner — Mega Hackathon
MegaHackLeague
-
2nd Runner Up — Codesphere Hackathon
Inter Campus Hackathon
-
Ocksket Scholarship
NEDUET
-
Gold Badge — Python
HackerRank
-
Maryam Faruqi Scholarship
Happy Home High School
-
SMEC'26 FYP Pitch Runner-Up
Sir Syed University
-
First Place at Tech Tank Probattle'26
IBA
-
Winner at AI Preneur’26
Fatimiyah College
-
Winners at Indus AI week
Sindh IT Company
-
First Runner-Up for Startup Showdown at Procom'26
FAST-NUCES
-
First Runner-Up for FYP Extreme at Procom'26
FAST-NUCES
Supervised Machine Learning: Regression & Classification — Coursera
Crash Course on Python — Coursera
Contact
Let's build something
worth talking about.
Open to internships, freelance projects, and full-time roles. If you're working on something interesting in AI, ML, or full-stack — reach out.