Hi, World! I'm Ashub Shafqat — pronounced Ash-ab 👋
I connect artificial intelligence, software engineering, research, and cybersecurity to build useful products—not just prototypes.
I am Ashub Shafqat, a Software Engineering graduate from the University of Agriculture Faisalabad, Pakistan. I enjoy building intelligent systems that solve practical problems in agriculture, research, industrial safety, cybersecurity, and decision support.
|
Agentic AI, Generative AI, RAG, multi-agent orchestration, LLM-powered applications, computer vision, voice AI, and intelligent decision-support systems. |
React, Next.js, TypeScript, Laravel, FastAPI, Flask, Supabase, PostgreSQL, MySQL, REST APIs, responsive interfaces, and production deployment. |
Windows endpoint security, Defender, firewall policies, PowerShell automation, secure AI workflows, human approval, remediation, and verification. |
class AshubShafqat:
def __init__(self):
self.roles = [
"AI Engineer",
"Full-Stack Developer",
"Researcher",
"Cybersecurity Enthusiast"
]
self.education = (
"BS Software Engineering — "
"University of Agriculture Faisalabad"
)
self.interests = [
"Agentic AI",
"Generative AI",
"Multi-Agent Systems",
"AI Security",
"Cybersecurity",
"Full-Stack Development",
"Applied Research"
]
self.building = [
"AgroReef",
"SentinelHeal AI",
"VC Mind",
"ResearchFlow AI"
]
self.philosophy = (
"Build intelligent systems that create real-world impact."
)
def current_focus(self):
return [
"AI agents with safe tool use",
"AI for agriculture",
"Cybersecurity automation",
"Research-driven product development",
"Open-source collaboration"
]- 🌾 Creator of AgroReef, an agentic AI agricultural advisory platform for Pakistani farmers
- 🥉 Mozilla Track — 3rd Place, Hack-Nation 4 Global AI Hackathon
- 🏆 National-level Generative AI Hackathon Winner
- 📖 Selected as an Associate Editor in Computer Science at Synthica Journal
- 🔬 Associate Researcher working on Edge AI for medical signal analysis
- 🔐 Cybersecurity internship experience with endpoint protection and Windows security
- 🌍 Built and presented AI systems in international hackathons and research communities
Agentic AI React TypeScript Supabase Gemini PWA
An agentic AI agricultural advisory platform designed for Pakistani farmers. AgroReef combines weather intelligence, crop-disease analysis, market awareness, voice interaction, multilingual support, field monitoring, alerts, and human-centered decision support.
Core capabilities
- Seven-agent architecture with Supervisor, Vision, Weather, Market, Critic, Reasoning, and Executive agents
- OPARD loop: Observe → Plan → Act → Reflect → Deliver
- Urdu and English voice interaction
- Crop disease detection through image analysis
- GPS-based weather and seven-day forecast
- Pakistani mandi market prices and sell/hold recommendations
- Field mapping, notebooks, alerts, and farmer-specific records
- Progressive Web App with secure Supabase authentication and Row-Level Security
🔗 Repository · Live Demo
Agentic AI Browser Automation Multi-Agent Systems
A browser-level AI assistant designed around permissioned actions, intelligent navigation, information extraction, decision-making, and safe execution.
The project earned 3rd place in the Mozilla Track at the Hack-Nation Global AI Hackathon.
Generative AI Full-Stack Investment Analysis
An AI system that helps analyze startups, investment opportunities, business signals, risks, market context, and venture-capital decisions through structured intelligence and an interactive interface.
Cybersecurity Flask Groq PowerShell
An AI-guided Windows endpoint inspection and remediation system with evidence-based verification.
Workflow
Inspect → Analyze → Approve → Execute → Verify
The system combines live endpoint evidence, AI-generated remediation, local security guardrails, human approval, auditable execution, and post-remediation verification.
Next.js AI Agents IoT Analytics Risk Scoring
An industrial monitoring dashboard for analyzing temperature, vibration, gas, and operational risks. It provides anomaly detection, bilingual insights, explainable risk scoring, and AI-assisted safety recommendations.
Streamlit FAISS Gemini Multi-Agent AI
A research-paper analysis system with specialized Reader, Reviewer, and Synthesizer agents. It supports semantic retrieval, structured review, evidence synthesis, and multi-document analysis.
IBM watsonx AI Agents Research Automation
A multi-agent research and email automation system developed for the IBM Dev Day “AI Demystified” Hackathon.
Machine Learning Data Analysis Dashboard
A machine-learning project focused on poverty prediction using socioeconomic data, feature engineering, predictive modeling, and data-driven visualization.
My research interests sit at the intersection of intelligent systems and real-world deployment:
- Agentic AI and autonomous decision systems
- Multi-agent orchestration
- Retrieval-Augmented Generation
- AI for agriculture and climate resilience
- AI security and safe tool use
- Edge AI for medical signal analysis
- Model compression and robustness
- Human-in-the-loop AI systems
- Generative AI evaluation
- Explainable AI and trustworthy automation
AGROREEF: An Agentic AI Agricultural Advisory System for Pakistani Farmers
A research-driven system exploring multi-agent reasoning, autonomous monitoring, multimodal interfaces, farmer-centered design, calibrated confidence, and agricultural decision support.
Edge AI for Real-Time Medical Signal Analysis
Research focused on how model compression affects latency, noise robustness, and cross-dataset generalization for edge-deployed ECG models.
- 🥉 Mozilla Track — 3rd Place, Hack-Nation Global AI Hackathon
- 🏆 National Generative AI Hackathon Winner
- 🌍 Top 16 Finalist among hundreds of global hackathon teams
- 📖 Associate Editor — Computer Science, Synthica Journal
- 🔬 Associate Researcher — Machine Learning & Applied Research, NEXEDGE Research Lab
- 🔐 Cybersecurity Intern, NETSOL Technologies
- 🌾 Built AgroReef as a research-backed final-year project
- 🧩 Participated in Harvard CS50x Puzzle Day
- 💡 Built solutions across agriculture, research, cybersecurity, industrial safety, and venture intelligence
This animation is generated automatically by the GitHub Actions workflow in this profile repository.
Contribution Snake GitHub Actions workflow
Create this file:
.github/workflows/snake.yml
Then paste:
name: Generate Contribution Snake
on:
schedule:
- cron: "0 0 * * *"
workflow_dispatch:
push:
branches:
- main
permissions:
contents: write
jobs:
generate:
runs-on: ubuntu-latest
timeout-minutes: 10
steps:
- name: Generate snake animation
uses: Platane/snk/svg-only@v3
with:
github_user_name: ${{ github.repository_owner }}
outputs: |
dist/github-contribution-grid-snake.svg
dist/github-contribution-grid-snake-dark.svg?palette=github-dark
- name: Publish snake animation
uses: crazy-max/ghaction-github-pages@v4
with:
build_dir: dist
target_branch: output
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}- Advanced agentic AI architectures
- Model Context Protocol
- AI security and safe autonomous agents
- Multi-agent collaboration
- Edge AI and efficient inference
- Research publication workflows
- Cybersecurity automation
- Cloud deployment and scalable AI systems
I am interested in:
- AI and Machine Learning roles
- Research collaborations
- Agentic AI projects
- Open-source contributions
- Cybersecurity and AI-security projects
- International hackathons
- Graduate research opportunities
- Technology for social impact
"Building intelligent systems that create real-world impact."