BUILDING & OPEN TO COLLABORATE // VADODARA, IN

I turn ideas into
software that solves
real problems.

I'm an Information Technology undergraduate at CHARUSAT exploring AI/ML, Computer Vision, Cybersecurity, and intelligent software products. Passionate about rapid experimentation, vibe coding, and translating ideas into functional engineering.

INSTITUTION CHARUSAT (DEPSTAR)
ACADEMIC CGPA 8.12 / 10.0
HACKATHON HONORS FinSpark '26 Top 20
PRIMARY TRACK AI/ML & Systems
Builder // 2025–29
Neel Agrawal — Information Technology Student & Developer
Neel Agrawal • Vadodara
LOC: 22.5996° N, 72.8205° E
[01] Selected Works & Active Prototypes

Engineered Systems & Applied AI

A transparent showcase of genuine projects, hackathon prototypes, and ongoing research. Every project is classified by actual progress and technical contribution.

Spotlight Project // Active Prototyping

Truthlence: Real-Time AI Deepfake Detection & Media Trust

An AI and computer-vision integrity platform engineered to uncover generative face-swaps, synthetic facial inconsistencies, and artifact anomalies in video streams in real time.

Neel's Role: Lead AI Vision Pipeline Architect (Frame ingestion, landmark normalization & FFT spectral analysis)
Python OpenCV TensorFlow NumPy MediaPipe 468 Mesh Digital Trust
SYSTEM TELEMETRY LATENCY: ~14MS/FRAME
VIDEO STREAM OpenCV Ingest FACE MESH 468 Temporal Jitter Track CONFIDENCE Anomaly Heatmap High-Frequency Fourier Spectral Artifact Trace
[02] Top 20 Team • FinSpark '26

Garuda AI: Insider Threat Detection Platform

A cybersecurity-focused machine learning project developed for FinSpark '26 that flags internal network compromise and abnormal data exfiltration using behavioral anomaly scoring.

Neel's Role: Data Pipeline, Feature Extraction & Anomaly Logic
Python NumPy Pandas Scikit-Learn Cybersecurity UEBA
[03] SIH Institute Selection

Anvesha: Autonomous Railway CCTV Safety AI

An AI-powered CCTV monitoring concept designed for an internal institute-level SIH problem statement, detecting crowd surges, platform bottlenecks, and security hazards across existing station cameras.

Neel's Role: Computer Vision Pipeline, Homography & Density Alert Logic
Python OpenCV Computer Vision RTSP Video Public Safety
[04] Hackathon Concept Prototype

Saarthi AI: Accessible Student Guidance

An AI-based counselling concept exploring confidential, stigma-free guidance for college students navigating academic workload stress, equipped with emergency helpline intercept triggers.

Neel's Role: Concept Architecture, Persona Design & Ethical Guardrails
Python Conversational AI NLP Triage Student Support
[05] Experimental Security Sandbox

Ghost Chat: Zero-Identity Ephemeral Chat

An experimental privacy-focused web chat prototype built around the principle of zero personal identifiers (no phone numbers, emails, or social IDs) and volatile RAM-only message retention.

Neel's Role: Full-Stack Prototype Developer (Ephemeral WebSockets)
JavaScript WebSockets HTML5 CSS3 Privacy Engineering
[06] ⚡ Currently Building

GPS-Based Context Timer

A location-aware automated timer combining geographic coordinate geofencing with time-based triggers to execute context-aware reminders and automated workflows upon entering designated physical zones.

Neel's Role: Lead Developer & Geofencing Pipeline
Python GPS / Geolocation Smart Automation
Repository Profile
[07] ⚡ Currently Building

Cloud Coding Workspace

A browser-based development environment enabling rapid coding experimentation in ephemeral cloud sandboxes without demanding local toolchain setup on user machines.

Neel's Role: Architect & Frontend Prototype Developer
Cloud Architecture Developer Tools Web IDE
Repository Profile
[02] Human Personality & Engineering Ethos

Vibe Coding Meets Rigorous Systems Thinking

I am an Information Technology student at Devang Patel Institute of Advance Technology and Research (DEPSTAR), CHARUSAT (2025–2029), maintaining a CGPA of 8.12.

"To me, vibe coding is not about cutting corners — it's about eliminating the friction between a creative spark and a working, testable system."

My journey began with raw curiosity about how software interacts with the physical and human world. Instead of simply studying algorithms from a textbook, I build them: extracting video frames to monitor crowd densities at railway stations, aggregating authorization logs to catch insider security threats, and testing deepfake detection pipelines on real video feeds.

I believe modern engineering thrives when deep technical discipline is paired with fast, fearless experimentation. Whether collaborating in high-pressure national hackathons like FinSpark '26 (where our team achieved a Top 20 national ranking) or crafting clean standalone tools, my focus remains constant: building practical technology that solves real human problems.

Based in Vadodara, Gujarat, I am fluent in English, Hindi, and Gujarati. I am currently deepening my knowledge in high-performance neural model inference, computer vision edge pipelines, and secure cloud system architecture.

01

Pragmatism Over Hype

I prioritize real utility over buzzwords. A simple heuristic with high precision is infinitely more valuable than an uninterpretable, bloated neural black box.

02

Security & Digital Trust First

Systems must be designed defensively from frame one. User privacy, credential integrity, and ethical data boundaries are non-negotiable architectural requirements.

03

High-Velocity Iteration

I prototype quickly to validate core assumptions with real data, then systematically harden code, eliminate race conditions, and polish user experience.

[03] Technical Toolkit

Languages, Frameworks & Stacks

Organized by verified practical experience across production projects, hackathons, and laboratory coursework. No arbitrary percentage progress bars.

Programming Languages

[05]
Python Active / Daily
C++ DSA / Academic
C Systems Core
Java OOP Principles
JavaScript (ES6+) Full-Stack / Web

AI, Vision & Data Science

[06]
TensorFlow Model Training
OpenCV Computer Vision
MediaPipe Landmark Tracking
NumPy Vector Math
Pandas Data Analytics
Scikit-Learn Anomaly Models

Web & Interface Engineering

[04]
HTML5 (Semantic) Accessible Core
Modern CSS3 Custom Design Systems
Responsive Systems Mobile to 4K
RESTful APIs Integration

Tools, Security & Workflow

[05]
Git & GitHub Version Control
Linux / CLI Shell Automation
Cybersecurity UEBA Threat Analytics
VS Code / Dev Tools Primary IDE
RTSP Video Feeds IP Stream Parsing

Human Communication

[03]
English Professional Working
Hindi Fluent Native
Gujarati Native Regional
[04] Verified Timeline

Academic & Hackathon Journey

A chronological timeline of substantiated educational qualifications, hackathon recognitions, and project milestones.

2026
🏆 National Hackathon Honor

Top 20 Finalist Team — FinSpark '26

National-Level Competitive Hackathon

Selected among the Top 20 teams nationwide as part of an engineering team developing Garuda AI — an intelligent cybersecurity platform designed to flag insider threats through anomalous behavioral scoring.

2025 – 2029
🎓 Undergraduate Degree

B.Tech in Information Technology

Devang Patel Institute of Advance Technology and Research (DEPSTAR), CHARUSAT

Pursuing comprehensive coursework in Data Structures, Algorithms, Computer Networks, Database Systems, Artificial Intelligence, and Computer Vision. Current Cumulative GPA: 8.12 / 10.0.

2025
⚙️ Innovation Competition

Smart India Hackathon (SIH) Internal Selection

CHARUSAT Internal Evaluations

Developed Anvesha as an AI-powered CCTV crowd monitoring solution for Indian Railways, qualifying through institute-level screening rounds with working computer vision demonstrations.

2025
🚀 Vibe Coding & Sandboxing

Hackathon Sprinting & Prototyping

Collaborative Developer Sprints

Prototyped Saarthi AI (conversational student guidance assistant) and Ghost Chat (ephemeral privacy chat), honing practical skills in rapid concept testing, UI design, and ethical guardrails.

[05] Connect & Collaborate

Let's build something
exceptional together.

I am open to hackathon collaborations, AI/ML research discussions, developer internships, and ambitious software projects. Reach out via email or direct profile links below.

Send a Direct Dispatch

Generates a formatted message dispatch to neelag0804@gmail.com with spam honeypot verification.