A studio site that sells the work before a word is read
Cinematic marketing site for a Nashik architecture practice — motion-led, image-heavy, built to make 145 completed projects feel like a portfolio worth commissioning.

- Projects shown
- 145+
- Sq. ft. designed
- 5,00,000+
- Years of practice
- 12+
- Client
- Pagar Architects
- Type
- Studio website
- Industry
- Architecture
- Delivered as
- Engineer at AI Alpha Tech
- Stack
- Next.js, React, TypeScript, GSAP, Tailwind CSS
The problem
An architecture practice is judged on presentation before credentials. The studio needed a site that carried the quality of the buildings themselves, and that made twelve years of practice and half a million square feet designed feel immediate rather than buried in a list.
What I built
A motion-led marketing site: a full-bleed hero with the practice's own renders, a numbered navigation system, service cards that read as an index, and a stat band putting projects, square footage and years of practice above the fold. Every animation is scroll-driven so the page reveals rather than loads.
What it does
- Full-bleed hero with scroll-driven reveals
- Numbered navigation and section indexing
- Service cards for residential, interior and commercial
- Above-the-fold credibility band
- Projects gallery and enquiry flow
The outcome
The studio leads with its work. The site sets the same standard as the architecture it represents, which is the entire job of a practice website.
Running the same problem?
Tell me what the process looks like today. I'll tell you honestly whether software fixes it.
More work
- ConstructionSiteLeaderMulti-tenant SaaS for Indian building contractors — labour, materials, finance, GST billing and AI-assisted design in one tenant-isolated platform.
- Society managementChairman Sahebn8n-built WhatsApp agent for the Chairman Saheb society app — answering questions, sending quotations, qualifying leads and booking demo slots against a live lead sheet.
- Voice AIChairman SahebReal-time voice agent for the Chairman Saheb app — the caller picks a language, and Sarvam speech-to-text, an OpenAI model and Sarvam's Shubh voice answer only from the app's knowledge base.