Last Look
In progressAn AI-assisted way to find local deals from people who need to sell.
Building Last Look.
An AI-assisted way to find local deals from people who need to sell.
Last Look started with a simple idea: people who need to sell something quickly are often willing to make a deal.
The problem is that those listings are scattered across different marketplaces, and finding them takes time.
I’m building Last Look to bring those opportunities together in one place. Using AI-assisted development tools, I’ve been building the frontend, backend, data pipelines, and automation needed to make it work.
Building the app and its data pipeline.
I started by defining the product, the types of listings I wanted to find, and what would make a seller look motivated.
From there, I built the backend in Replit, connected it to a Supabase database, and used Apify to collect listings from different marketplaces. I used ChatGPT throughout development to help plan features, write and debug code, and work through technical problems. The consumer-facing frontend is built with Lovable, and PostHog helps me understand how people use the site.
Collecting data
- Automated listing ingestion
- Public listing feeds
Cleaning data
- Listing cleanup
- Duplicate detection
- Freshness & expiration filtering
Understanding listings
- Geographic classification
- Location filtering
- Seller-motivation scoring
The app
- Consumer frontend
- Protected admin endpoints
Still in active development. Some marketplace integrations are more reliable than others.
From scattered listings to one feed.
- 1
Collect
Gather listings from supported marketplaces.
- 2
Clean
Normalize listing information and remove duplicates.
- 3
Evaluate
Identify seller-motivation signals and time-sensitive opportunities.
- 4
Filter
Organize listings by location, freshness, and other criteria.
- 5
Discover
Show relevant listings in the Last Look app.
The tools behind it.
Replit
Application development, backend APIs, data ingestion workflows, and deployment.
Supabase
PostgreSQL database that stores, organizes, and manages marketplace listings.
Apify
Automated marketplace data collection, including Facebook Marketplace and OfferUp.
ChatGPT
Planning, technical architecture, coding, debugging, testing, and iteration.
Lovable
Frontend development and interface design for the consumer app.
PostHog
Product analytics and website traffic analysis.
What I’ve learned building it.
Building Last Look has pushed me to work on parts of a product I wouldn’t normally own in a product marketing role.
I’ve been working through database design, API development, data quality, automation, and frontend development.
AI tools have made it possible for me to take on more of that work myself, but there’s still plenty of debugging, testing, and figuring things out along the way.
It’s been a good reminder that building something is one thing. Making it useful and reliable is another.
Additional Last Look screenshots coming soon.
More good products.
Let’s connect.
I’m currently exploring my next full-time opportunity in product marketing, growth, or product strategy.
If you’re hiring, building an interesting product, or think there might be a fit, I’d love to connect.