Nicheloom

Market intelligence for builders — see what's gaining traction before it's crowded.

See what readers who loved your favorite book/author also loved to read

Details

External ID
46419822
Source
HN
Company
—
Product
See what readers who loved your favorite book/author also loved to read
Website domain
shepherd.com
Launched
Dec. 29, 2025
Cohort
—
Upvotes
134
Upvotes percentile
0.9179389312977099
Tags
—
Fetched at
Sept. 7, 2026, 9:25 p.m.
Updated at
Sept. 7, 2026, 9:25 p.m.

Description

Hi HN,Every year, we ask thousands of readers (and authors) to share their 3 favorite reads of the year.Now you can enter a book/author you love and see what books readers loved who also loved that book/author.Try it here: https://shepherd.com/bboy/2025This goes wide and doesn't try to limit itself to the genre, so you get some interesting results.What do you think?Background:I want better recommendations based on my reading history. I'm incredibly frustrated with what is out there.This system is based on 5,000 readers voting on their 3 favorite reads from 2023 to 2025. So, this covers ~15,000 books and is a high-quality vote. We wanted to keep the dataset small for now while we play with approaches.We are building a full Book DNA app that pulls in your Goodreads history and delivers deeply personalized book recommendations based on people who like similar books (a significant challenge).You can sign up to beta test it here if you want to help me with that:https://docs.google.com/forms/d/1VOm8XOMU0ygMSTSKi9F0nExnGwo...The first beta is coming out in late January, but it's pretty basic to start.Past Show HNs as we've built Shepherd:https://news.ycombinator.com/item?id=40084193https://news.ycombinator.com/item?id=38600246https://news.ycombinator.com/item?id=26871660Thanks, looking forward to your comments :)Ben

Enrichment

Theme
media discovery and streaming tools
Vertical
Media & entertainment
Function
Search & retrieval
Audience
B2C
AI stance
Not AI
Project type
Hobby / open-source project
Normalized one-liner
book recommendations based on reading history
Manually corrected
False

Could you build this?

Partial The recommendation frontend and simple collaborative filtering matching can easily be vibe-coded, but the value is gated by a proprietary curated dataset of thousands of human and author recommendations.

What it would actually take: A complete system needs a relational database (PostgreSQL) storing graph relationships between readers, authors, and books, along with a collaborative filtering or bipartite matching algorithm. The UI can be built with Next.js or Remix. The primary bottleneck is proprietary crowdsourced data acquisition—reaching thousands of real readers and vetted authors to build the preference graph.

Discussion

20 comments analyzed.

Competitors mentioned: LLMs as universal recommendation engines, Hacker News (HN) for trust/voting models

Concerns raised: Search results unclear whether algorithmic or handpicked lists, Gender-based recommendations inappropriate (only female authors shown), Older books missing from database, Author fraud via fake accounts voting for own books, Mobile UX hard to find book recommendation lists

Feature requests: Account creation without email requirement, Button to access 'Books Like X' directly from book page, Surface recommendation lists outside collapsibles on mobile, Use negative ratings in recommendations, Add older/classic books like Dreamsnake to database

Competitors

Other products that read as similar to this one — 94 launches clear the similarity bar, closest 8 shown.

Attention rank: #12 of 95 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).

Launched 54 days after the earliest competitor.

Other launches for this product