AI/ML Aspirant · Student Builder
I’m Priyanshu Zihowa, an AI/ML student from Lucknow, exploring AI, machine learning, and thoughtful digital products.
Priyanshu Zihowa
About
I’m an AI/ML student from Lucknow. I spend my time learning how artificial intelligence and machine learning actually work, and turning what I learn into small, real projects.
I’m at the beginning of this path, and I’m honest about that. What I bring is curiosity: about how AI actually works, and about how it can be used to make everyday things genuinely better for people.
I learn best by building. Instead of only reading about ideas, I try to turn them into something real, like Lumo AI, an early project exploring how AI can help students study. Every attempt teaches me something new, and I want to keep creating technology that is useful, not just impressive.
Steadily learning the ideas behind artificial intelligence and machine learning, and where they are truly useful.
Turning curiosity into small, real projects and improving them one iteration at a time.
Staying open, adaptable, and willing to keep learning as the field keeps changing.
Featured project
An AI-powered study companion for students.
Lumo AI is designed to make studying more interactive and accessible for students in Classes 9–12 and learners preparing for competitive exams such as JEE and NEET. The product explores how AI can support focused learning, clearer explanations, and better exam preparation.
Exam mode
Focused revision for JEE & NEET.
Today
What would you like to understand today?
Physics
Motion & forces
Chemistry
Chemical bonding
Mathematics
Calculus basics
Biology
Cell structure
68%
Focus
Today’s focus
3 topics revised, 1 to go.
Up next
Revise: Newton’s laws of motion
Ask Lumo
Your AI study companion
Illustrative concept preview, designed in HTML & CSS for this portfolio.
Data Engineering · Machine Learning · Web · Live · growing
A platform for understanding Dubai's property market.
A web platform built on 105,986 official Dubai Land Department sales: a market dashboard, a page for each of 122 areas, like-for-like market insights, a machine-learning price estimator (8.3% median error on unseen sales) and an AI assistant that answers questions by calling tools over the data.
AI · RAG · Legal Tech · Live
Ask about your rights at work, get answers that cite the law.
A RAG assistant for the UAE Labour Law: ask in plain words (“what notice period do I give if I resign?”) and get an answer that cites and links the exact articles. Search finds the right article in the top 3 for 90% of unseen test questions. With an exact gratuity calculator.
Mobile App · AgriTech · Live demo · v0.15
Book farm machines like you book a ride.
A Hindi-first Android app that connects farmers in Uttar Pradesh with tractor and farm-machine owners and their drivers: book a machine by voice or tap, track it to the field on a map, and pay by UPI. Built with Flutter and a Supabase backend.
Deep Learning · AgriTech · Live demo
Spot crop disease from a photo of a leaf.
Photograph a potato, tomato or maize leaf and a neural network names which of 17 diseases (or healthy) it looks like, with simple advice in Hindi. 97.1% accurate on leaves it never saw, and it runs on the phone, so the photo is never uploaded. Part of KhetGo.
Computer Vision · Public Safety · Live demo
Privacy-first people tracking and safety alerts, in the browser.
Detects people in a live camera or video, follows each one with an ID using a tracker I wrote, and raises alerts for restricted zones, loitering and crowds. Everything runs on the device, with an optional privacy blur.
Deep Learning · Computer Vision · Live demo
Draw a Devanagari character, and a neural network recognises it.
A convolutional neural network trained on 78,200 handwritten Devanagari characters. It recognises 46 characters (36 consonants and 10 digits) with 99.6% accuracy on unseen test images, and runs entirely in the browser.
Vision
A simple roadmap: finish strong, learn deeply, then build things that genuinely help people.
Now
01Studying the fundamentals of AI and machine learning, and building projects like Lumo AI along the way.
Next
02Building strong foundations in mathematics, programming, and how intelligent systems work, and learning from real-world experience.
Ahead
03Turning what I learn into useful tools, especially ones that make learning clearer and more accessible.
Interests
Areas I’m actively learning about and experimenting with.
“I believe strong AI products begin with useful problems, clear thinking, and a willingness to keep learning.”
Coming soon
Ideas I’m planning to explore next. These are upcoming explorations, not finished work.
A space for small experiments with AI tools and ideas as I keep learning. Some may grow into projects; all of them will be lessons.
Idea · Upcoming exploration
Contact
I’m open to learning opportunities, collaborations, student programs, and conversations around AI and technology.
