Hacker News

Is this an idea worth pursuing or not?

Hacker News - Mon, 11/04/2024 - 3:01pm

The social scene in big cities like New York or LA can be daunting. Drinks are expensive, places can be crowded, lines long and the service sucks. So I was thinking, what if you have a platform that enables people to turn ordinary private spaces such as their home (or really anywhere) into social settings? For instance, turn your apartment into a bar with 4 tables and guests can reserve a spot (either with a group of friends or to hang out with strangers) by paying an admission fee (will be byob to avoid liquor license issues). This concept can be applied to different social settings such as hosting a mom’s night for other moms with kid care included, hosting a mediterranean dinner party, hosting a games night etc. The platform would enable guests to see what events are happening in their immediate vicinity / or a particular location on a specific date and reserve spots at various events they want to attend. Guests can also see if their friends are attending a particular event (if the guest makes their attendance public) and over time see if any guests that they’ve met previously will be attending. Guests and hosts can leave reviews on each other's profiles as feedback for future users. Some immediate issues I’ve identified are the usual market place chicken and egg problem and potential security concerns about letting strangers into your private space (though the same was true for uber and airbnb). Are there any other potential issues? I have several ideas on how to monetize it but wanted to hear what people think about the concept.

Comments URL: https://news.ycombinator.com/item?id=42045469

Points: 1

# Comments: 4

Categories: Hacker News

Down in the Mantle

Hacker News - Mon, 11/04/2024 - 2:16pm
Categories: Hacker News

Show HN: Fuzzy deduplicate any CSV using vector embeddings

Hacker News - Mon, 11/04/2024 - 2:07pm

I made an app to fuzzy-deduplicate my Google Sheets and CRM records

- No manual configuration required

- Works out-of-the-box on most data types (ex. people, companies, product catalog)

Implementation details:

- Embeds records using an E5-family model

- Performs similarity search using DuckDB w/ vector similarity extension

- Does last-mile comparison and merges duplicates using Claude

Demo video: https://youtu.be/7mZ0kdwXBwM

Github repo (Apache 2.0 licensed): https://github.com/SnowPilotOrg/dedupe_it

Background story: My company has a table for tracking leads, which includes website visitors, demo form submissions, app signups, and manual entries. It’s full of duplicates. And writing formulas to merge those dupes has been a massive PITA.

I figured that an LLM could handle any data shape and give me a way to deal with tricky custom rules like “treat international subsidiaries as distinct from their parent company”.

The challenging thing was avoiding an NxN comparison matrix. The solution I came up with was first narrowing down our search space using vector embeddings + semantic similarity search, and then using a generative LLM only to compare a few nearest neighbors and merge.

Some cool attributes of this approach:

- Can work incrementally (no reprocessing the entire dataset)

- Allows processing all records in parallel

- Composes with deterministic dedupe rules

Lmk any feedback on how to make this better!

Comments URL: https://news.ycombinator.com/item?id=42044962

Points: 2

# Comments: 0

Categories: Hacker News

Shaping tools that shape us at Notion

Hacker News - Mon, 11/04/2024 - 2:02pm

Article URL: https://jwn.gr/posts/joining-notion/

Comments URL: https://news.ycombinator.com/item?id=42044916

Points: 1

# Comments: 0

Categories: Hacker News

Pages