←Back/Research/Claude/ChatGPT
AdvancedResearch

How to Streamline Archival Research with a Custom AI-Powered iOS App

Build a field research app that captures the front and back of an archival photo as a linked pair, generates a description and transcription immediately, embeds both into image metadata, and gives every file a predictable name.

How to Streamline Archival Research with a Custom AI-Powered iOS App

Source episode · 24:21

How I AI: Tim McAleer's AI Workflows for Documentary Filmmaking at Florentine Films

with Claire Vo

Tim demos Flip Flop by photographing the front and handwritten back of a print, then shows the generated description, transcription, EXIF metadata, and structured filenames.

Read the full episodeWatch on YouTubeListen on SpotifyListen on Apple

Before you start

What you need

  • An iPhone and Xcode
  • A vision and OCR API
  • A naming convention for collections and front or back pairs
  • A destination that preserves image metadata

What you’ll make

An iOS capture app that produces linked, consistently named image pairs with descriptions and transcriptions embedded in their metadata.

Tools used

Step by step

The workflow

Follow the sequence once, then adapt the prompts, checks, and handoffs to your own setup.

5 steps

Step01

Define App Requirements with an LLM

Write the app brief around the real archive sequence: create a collection, capture a front, optionally add notes, capture the matching back, process both, and export predictable files. Define offline and retry behavior before generating Swift code.

Example prompt
Write a concise PRD for an iOS archive capture app called Flip Flop. A researcher creates a collection, photographs the front of an item, immediately photographs its back, adds optional notes, and continues to the next item. Define the pairing identifier, filename pattern, processing states, offline queue, metadata fields, and export flow.
Step02

Generate UI Code with an AI Assistant

Use the PRD to generate the SwiftUI shell with collection management, camera capture, pair review, processing status, and export. Keep capture fast enough for a researcher working through hundreds of items.

Example prompt
Implement the SwiftUI foundation from the attached PRD. Include collection creation, front and back capture, optional notes, a persistent pair record, background processing status, retry, and export. Use a simple camera first interface and keep services separated from views.
Step03

Implement the Core Capture Workflow

When the front is captured, create the item ID and prompt for the back without leaving the flow. Save both originals locally with the same identifier, capture order, and front or back suffix before starting any network request.

Step04

Integrate AI for Description and Transcription

Send the front to the vision model for a factual description and the back to OCR for transcription. Include notes and capture context, preserve uncertainty, and write the returned text to the pair record.

Step05

Embed AI-Generated Data into Image Metadata

Embed the description, transcription, collection, item ID, capture date, and notes into standard image metadata fields. Export the originals with metadata intact and verify the fields and filenames on another computer.

What good looks like

  • The front and back remain paired after export.
  • Descriptions and transcriptions appear in readable image metadata fields.
  • Files sort predictably by collection and capture order.
  • A researcher can capture many items without returning to the camera roll.

Build your next product with ChatPRD

Turn an idea into a PRD, user stories, and a plan.

Try ChatPRD free

After the steps

Runbook notes

How to recover when the loop fails and where human judgment helps.

Recover

If it goes sideways

The front and back become separated
Create the pair record before capture and assign both files the same collection and item identifier with explicit front and back suffixes.
Export strips the generated metadata
Write standard EXIF, IPTC, or XMP fields to the final file and test the actual transfer method with an independent metadata reader.
The back transcription is uncertain
Keep the source image, mark uncertain words, and allow a note or correction before final export.
Network processing interrupts rapid capture
Save the pair locally first, process it in a background queue, and show pending, complete, and failed states without blocking the next capture.

Keep building

Related workflows

Create a Meta-Workflow to Continuously Improve Your AI Assistant's Performance
advancedPersonalClaude

Create a Meta-Workflow to Continuously Improve Your AI Assistant's Performance

Design a weekly meta-workflow that analyzes your interactions with your AI assistant to make it smarter. This system learns your writing style from edits, suggests new automations for repetitive tasks, and helps you filter valuable AI techniques from hype.

Open
Build a Self-Improving AI Morning Brief to Capture Action Items and Learn Company Jargon
intermediateProductClaudeGranola

Build a Self-Improving AI Morning Brief to Capture Action Items and Learn Company Jargon

Create a daily AI workflow that summarizes yesterday's meetings, identifies urgent tasks, and proactively learns internal company terms it doesn't understand. This keeps your AI assistant's knowledge current and highly relevant to your work.

Open
Automate Your Weekly Planning with an AI-Powered PM Assistant
intermediateProductClaudeNotion+1

Automate Your Weekly Planning with an AI-Powered PM Assistant

Build a recurring AI task that automatically gathers updates from your calendar, Slack, and notes to suggest weekly priorities and prepare you for upcoming meetings. Start every Monday with a clear, comprehensive plan without the manual effort.

Open

Start shipping
better products.

Join 100,000+ product managers who use ChatPRD to write better docs, align teams faster, and build products users love.

Start building freeRequest a demo
Free to start
No credit card
SOC 2 certified
Enterprise ready