Project Proposal · Draft

A guest walks in, and their photographs find their way home.

An end-to-end platform for K.Nasif Photography: guests are enrolled at the door, faces are matched to the wedding gallery afterward, and every person automatically receives the photos they appear in — plus a branded, spacious home for years of client galleries.

Prepared for
K.Nasif Photography
Prepared by
Unityflow AI
Date
17 July 2026
Status
Draft estimate
2
Products in one platform
~7
Weeks to first live wedding
8
Delivery phases
TB+
Storage, priced simply
01

What we're building

Two connected products under one roof. The first delivers photos to the right guests automatically. The second gives the studio a branded, generous place to keep every gallery — the kind of storage clients already understand, without the technical fine print.

Guest Photo Delivery

At the event, each guest is enrolled once — a quick reference photo plus a way to reach them (a scannable QR code, or their email on a sign-in sheet). After the wedding, the system finds every photo that guest appears in and sends their personal gallery straight to them.

Branded Photo Storage

A private, Google-Drive-style space for the studio's galleries — upload, organize, and share terabytes of photos with clients. Clean monthly pricing per gigabyte, no jargon about bandwidth or transfer fees. The studio always knows exactly what a gallery costs.

02

How the photo delivery works

Four moments, from the guest arriving to the photos landing in their inbox. The studio only ever does two things: enroll guests at the door, and upload the gallery afterward. The rest runs on its own.

STEP 01

Welcome & enroll

At the entrance, a guest is captured once. They either get a QR code tied to their reference photo, or write their email on a sign-in form. Takes seconds per person.

STEP 02

Shoot the day

The photographer works the wedding exactly as always. Nothing changes about how they shoot — the platform stays out of the way.

STEP 03

Upload & match

Afterward, the full gallery is uploaded. The system recognizes faces and matches each guest to every photo they appear in, grouping a personal set per person.

STEP 04

Deliver automatically

Each guest receives their own gallery — by email, or by scanning their QR code. No manual sorting, no "which photos are mine?" from a hundred guests.

On accuracy & consent. Face matching is assistive, not perfect — a quick review step lets the studio approve borderline matches before anything is sent, so no guest receives a stranger's photos. Guests opt in when they enroll, and can request removal at any time. This keeps the experience delightful and the studio's reputation protected.
03

What's inside the platform

Built for a working studio, not a demo. Every part is designed to run one wedding after another without hand-holding.

Two ways to enroll

QR code on arrival for tech-comfortable guests, or a simple email sign-in sheet as a fallback. The couple's older relatives are covered too.

Face matching

Each uploaded photo is scanned and matched against enrolled guests, building a personal gallery per person automatically.

Review before send

An approval view for uncertain matches, so the studio stays in control and every delivery is one it's happy to put its name on.

Automated delivery

Personal galleries go out by email or QR link, on the studio's schedule — one wedding or a whole season's backlog.

Event dashboard

One place per wedding: guests enrolled, photos matched, galleries sent, and anything still needing a look.

Built to hold up

Rate limiting, error handling, and an automated test suite so a busy Saturday upload doesn't fall over. Reliability is part of the build, not an afterthought.

04

Storage, priced the simple way

The studio pays one clear rate per gigabyte, per month — that's it. No "ingress," no "egress," no surprise bandwidth bills when a client downloads their whole gallery. Below are illustrative monthly plans; final numbers are set once expected volume is confirmed.

Starter
250 GB
$15 / month
≈ $0.06 per GB · a few weddings live at once
  • Branded upload & share portal
  • Client gallery links
  • Unlimited downloads — no transfer fees
Archive
5 TB+
$180 / month
≈ $0.036 per GB · years of work, kept safe
  • Everything in Studio
  • Best per-GB rate, scales with you
  • Custom retention & access rules
05

Scope & effort

Eight phases from kickoff to handover, billed at the agreed hourly rate.

PhaseKey deliverablesHours
1 · Discovery & designRequirements, data model, architecture, environment setup16
2 · Core foundationAuth, roles, API skeleton, database schema, admin shell28
3 · Guest enrollmentQR issuing + reference capture, email sign-in flow, guest records30
4 · Face recognition pipelineDetection, embedding, matching, per-guest gallery grouping50
5 · Review & deliveryMatch approval UI, email/QR delivery, per-event dashboard36
6 · Storage portalBranded upload & share portal, client gallery links, backups44
7 · Testing, QA & hardeningTest suite, rate limiting, error handling, security review28
8 · Deployment & handoverCloud deploy, documentation, walkthrough, support setup18
Total estimated effort250 h
06

Indicative timeline

Roughly seven weeks at a steady pace, with phases overlapping where it's practical. The goal is a first real wedding running end-to-end as early as possible.

Week 1

Discovery & foundation

Lock requirements, stand up the core app, auth, and data model.

Week 2–3

Enrollment & pipeline

Guest sign-in (QR + email) and the face-matching engine take shape.

Week 3–4

Review & delivery

Approval flow, automated galleries, and the per-event dashboard.

Week 4–5

Storage portal

Build the branded upload and share portal.

Week 6

Testing & hardening

Test suite, rate limiting, error handling, security review.

Week 7

Deploy & handover

Go live, documentation, and a walkthrough with the studio.

07

Technology

Modern, proven, and open-source-first — chosen for reliability and no vendor lock-in.

Frontend

Next.js TypeScript Tailwind Automated tests

Backend

Python · FastAPI REST API layer PostgreSQL Auth & roles

Recognition

Face detection Face embeddings Vector matching Review queue

Storage & infra

Object storage No transfer fees Rate limiting Backups & monitoring
08

Assumptions & notes