I am a Computer Science Engineering student at Lovely Professional University with a 9.66 CGPA and recognition on the University Honor Roll. I pursue an AI/ML Minor alongside core CSE coursework, and I build software that runs in the real world: deployed ML models with 93%+ accuracy, production React applications used by students, hardware-software systems with native Apple HomeKit integration, and AI-powered full-stack applications with facial recognition.
My engineering work spans the full stack, from ESP8266 sensors streaming live data into FastAPI backends, to ML pipelines deployed on Hugging Face, to interactive web apps shipped on Vercel. I cleared the JPMorgan Chase Code For Good HackerRank round and contribute to open source through GSSoC '26.
I solve DSA problems in C++ (Striver A2Z roadmap), build AI/ML systems in Python, and develop web applications with React, FastAPI, and Node.js. My goal is to combine strong computer science fundamentals with practical AI/ML capabilities to build intelligent, production-quality software.
Open to: SWE Internships · AI/ML Roles · Open Source Collaborations
| Domain | Details |
|---|---|
| Machine Learning | RandomForest, GridSearchCV, cross-validation, OOB scoring, feature importance analysis |
| Applied ML + IoT | Sensor-to-model pipelines, real-time inference (27ms avg latency), zero train-deploy feature mismatch |
| Deep Learning | CNNs, MobileNetV2 (applied), PyTorch fundamentals, AI/ML Minor coursework at LPU |
| Computer Vision | OpenCV, dlib face recognition, TCS3200 color sensing, ESP32-CAM integration |
| NLP & Generative AI | Transformers, BERT, GPT concepts, OCI AI services, in-context learning, RAG foundations |
| Data Science | Pandas, NumPy, Matplotlib, EDA, hypothesis testing, probability and statistics |
| Cloud AI | Oracle Cloud Infrastructure AI Foundations (Certified), Hugging Face Spaces deployment |
🌿 Smart Crop Intelligence System: IoT + ML Precision Agriculture Platform
Real-time crop recommendation and plant disease detection system combining edge IoT sensors with a deployed ML backend.
| Attribute | Detail |
|---|---|
| ML Model | RandomForestClassifier, 300 estimators, GridSearchCV tuned |
| Accuracy | 93.18% test accuracy, 93.10% macro F1 score |
| Validation | 5-fold CV mean 91.19%, OOB score 91.82% |
| Dataset | 2,200 balanced samples across 22 crop classes |
| Feature Importance | Rainfall 0.38, Humidity 0.38, Temperature 0.24 |
| Inference Speed | 27.61ms average end-to-end latency |
| IoT Hardware | ESP8266 NodeMCU, DHT11, YL-83 rain sensor, soil moisture probe |
| Disease Engine | Rule-based evaluation of 11 environmental conditions across 4 severity levels |
| Stack | Python, FastAPI, Docker, Hugging Face Spaces, Chart.js |
| Endpoints | /predict (dashboard), /predict/compact (memory-optimized for MCU) |
| Course | CSE274 Applied Machine Learning at LPU |
| Repository | smart-crop-intelligence-system |
| Live Demo | fayasx-smart-plant.hf.space |
Engineering highlight: The model is trained exclusively on sensor-measurable variables (no NPK or pH inputs that the hardware cannot read), ensuring zero train-deploy feature mismatch. Rolling telemetry is visualized live in the web dashboard.
⚙️ Cpu Sched: Interactive CPU Scheduling Simulator
Production-grade Single Page Application that transforms abstract OS scheduling theory into dynamic, interactive learning.
| Attribute | Detail |
|---|---|
| Stack | React 18, Vite, Tailwind CSS, Zustand, Lucide React |
| Algorithms | FCFS, SJF, SRTF, Round Robin, Priority Scheduling |
| Features | Animated Gantt charts, Compare All mode, Queue Monitor, State Transition Diagram |
| Metrics | Auto-computed Waiting Time, Turnaround Time, Response Time, CPU Utilization |
| Architecture | Unidirectional data flow with pure scheduling functions decoupled from UI, Zustand as single source of truth |
| Repository | cpuSched |
| Live URL | cpu-sched.vercel.app |
| License | MIT Open Source |
| Team | Fayas Khan, Aditya Rana, Adithyan P (Mentor: Rizwan Sir) |
Engineering highlight: Students can configure arbitrary process workloads and watch all five major scheduling algorithms execute in real time with play/pause/step controls, timestamped logs, built-in educational scenarios (Convoy Effect, Starvation, Priority Inversion), and contextual algorithm explainers.
🏠 Smart Home Automation Socket: ESP8266 + Apple HomeKit Integration
Native Apple HomeKit smart home controller running a full HAP (HomeKit Accessory Protocol) server on an ESP8266 microcontroller, with no cloud dependency, no bridge device, and no subscription.
| Attribute | Detail |
|---|---|
| Stack | C++ (80 lines), ESP8266 NodeMCU, Dual Relay Module |
| Protocol | HAP (HomeKit Accessory Protocol) running entirely on-chip |
| Integration | Native iPhone, iPad, Mac, Apple Watch, HomePod support via Siri |
| Safety | Active-LOW relay logic (both outputs default to OFF on power-up) |
| Architecture | Wi-Fi, protocol handling, pairing, and relay control all on a single microcontroller |
| Repository | SmartHomeAutomationSocket |
Engineering highlight: The ESP8266 runs a full HAP server on-chip, handling Wi-Fi, protocol negotiation, device pairing, and relay control simultaneously. No Raspberry Pi or external hub required. Three source files, eighty lines of C++, full Apple ecosystem integration.
👨🏻💻 Face Attendance System: AI-Powered Recognition + Geotagging
Advanced attendance system using facial recognition and geotagging for secure, automated tracking with a high-performance backend and real-time face detection.
| Attribute | Detail |
|---|---|
| Stack | Python, FastAPI, React, Tailwind CSS, SQLite, dlib |
| AI/ML | dlib face recognition, real-time face detection |
| Features | Geotagging, automated tracking, real-time detection dashboard |
| Architecture | FastAPI backend with React frontend, computer vision inference pipeline |
| Repository | Face-Attendance-System |
| Live Demo | studentsfaceattendancesystem.vercel.app |
Engineering highlight: Combines computer vision (dlib face recognition), geolocation services, and a full-stack web application into a cohesive attendance management system. Demonstrates integration of AI inference with production web architecture.
| Platform | Status |
|---|---|
| GSSoC '26 | Accepted Contributor, GirlScript Summer of Code, Open Source Track |
| JPMorgan Chase Code For Good | HackerRank Coding Round Cleared, HireVue Invitation Received (SEP 2027) |
| Web-A-Thon Hackathon | Built "Budget Tracker for Students" using ReactJS + NodeJS |
| Hack-A-Throne 1.0 (AIESEC) | Frontend development, PC Building web platform |
| Recognition | Details |
|---|---|
| LPU University Honor Roll | Awarded for outstanding academic performance |
| CGPA 9.66 | Consistent excellence: S1 9.55, S2 9.63, S3 9.76, S4 9.72 |
| JPMorgan Chase Code For Good | Cleared HackerRank round, received HireVue interview invitation |
| GSSoC '26 Contributor | Accepted into GirlScript Summer of Code 2026 |
| 93.18% ML Accuracy | Deployed production crop intelligence model on Hugging Face |
| CPUSCHED Live Deploy | Shipped open-source React app used by OS students at LPU |
| Inspire Manak Fair | Science fair participation and project demonstration |
Oracle
NASSCOM
Infosys
Quickdraw
|
YOLO
|
PullShark
|
| Semester | TGPA | Key Subjects |
|---|---|---|
| Semester 1 | 9.55 | Python, Computing, Engineering Mathematics |
| Semester 2 | 9.63 | DBMS, Software Engineering, Computer Programming, Discrete Mathematics |
| Semester 3 | 9.76 | OOP, Data Structures and Algorithms, Computer Networks, Probability and Statistics |
| Semester 4 | 9.72 | Operating Systems, Design and Analysis of Algorithms, Theory of Computation, Computer Organization |
| Overall CGPA | 9.66 | University Honor Roll, AI/ML Minor |
Primary Track: C++ DSA
Roadmap: Striver A2Z (sequential, no skipping)
Language: C++ (primary problem-solving language)
Practice: 25-minute attempt rule before any hints
Goal: 200+ cold-solved problems by Semester 6
Secondary Track: AI/ML
Minor: AI/ML Minor at LPU
Domain: Supervised Learning, Deep Learning, NLP, Computer Vision
Focus: End-to-end projects with real deployment
Building:
- AI/ML projects with production deployment pipelines
- Open source contributions via GSSoC '26
- Backend APIs and intelligent applications
Open To:
- Software Engineering internships at product companies
- AI/ML Engineer opportunities
- Open source collaborations on impactful projects
Building intelligent software at the intersection of AI/ML, full-stack engineering, and embedded systems.