Hacker News

Closed Beta for a Smart Dating Assistant – 15 Spots Available

Hacker News - Thu, 11/14/2024 - 1:16pm

Hey hackers

We’ve been developing an AI-powered dating coach, trained on 30k+ real dating interactions from our matchmaking service. The goal? Help guys level up their dating game with better profiles, messages that actually get replies, and advice right when they need it.

We’re opening up just 15 spots for early users to join our private community, test features, and shape the product. In return for your honest feedback, you’ll get 3 months of free access (normally $40/month).

Who we’re looking for:

Men 18–30 Active on Tinder or Bumble Based in English-speaking regions

Here’s the link: https://bit.ly/beta_ws_vip_invite

It’s super quick—should take just a minute. Let me know once you’re done! Let’s build something great together!

Comments URL: https://news.ycombinator.com/item?id=42139288

Points: 1

# Comments: 0

Categories: Hacker News

How we built a new type of video search

Hacker News - Thu, 11/14/2024 - 1:15pm

Hey HN, we recently launched Remy, an AI agent designed to take the pain out of video search, and wanted to give the HN community a technical deep dive on how it works under the hood.

The Problem: There’s a ton of valuable content on the internet, but finding the “best parts” of long videos is frustratingly inefficient. Current video search methods haven’t evolved much since the early days of YouTube and aren’t designed for today’s massive volume and variety of information. Instead, we’re left scrubbing through long videos or, worse, missing valuable content entirely due to decision paralysis. Remy’s goal is to offer a smarter way to surface exactly what you’re looking for, in a fraction of the time.

The Solution: In less than a minute, Remy delivers custom playlists that isolate the best video moments from across the internet. It finds, clips, and organises segments to get you exactly what you need — without the endless search and skip game.

How it works:

Remy is powered by a stack of LLMs (and some non-LLM magic) designed for fast, focused video search and transcript processing. Here’s the pipeline:

1. Request Analysis When you send a message, the system decides whether to provide an immediate response, search the web, or start assembling video clips. If video is the best option, Remy generates a playlist outline with concise titles and detailed descriptions tailored to your query. For temporal queries, Remy automatically adjusts context to absolute dates (e.g., “tomorrow’s NBA games” → “November 15th NBA Games”).

2. Content Retrieval Using a ‘wide net’ approach, Remy generates a large set of targeted queries and searches the web for videos that could match your needs.

3. Multi-Step Filtration & Processing Each video goes through:

- Transcript Pull: Extracts YouTube transcripts.

- Non-LLM Filter: Filters out low-quality or AI-generated content based on YouTube stats, creator channels, release date, and other parameters.

- Punctuation Restoration (BERT): Restores punctuation for better LLM comprehension.

- Clipping: Uses an LLM to locate and clip the most relevant segments of the transcript based on your request.

- Evaluation: Uses an LLM to score clips on relevance, completeness, and interest level. Only the best make it to the final playlist.

4. Reordering & Overview A final LLM gets fed the top 16 clips for each topic, filters them down to the best 4 (at most), and arranges them for maximum uniqueness. Each section gets a brief overview, with added context pulled and cited from the web to give you additional context.

The result? A playlist tailored to your exact query, delivered in under a minute, without unnecessary noise.

We’d love to hear your thoughts and feedback. If there’s something you’re curious to try or any edge cases that it's not handling, let us know. And if you come across any bad clips, please use the report button to flag them! Thanks

https://www.useremy.com/

Comments URL: https://news.ycombinator.com/item?id=42139277

Points: 2

# Comments: 0

Categories: Hacker News

Ask HN: Why are drones invented so lately

Hacker News - Thu, 11/14/2024 - 1:14pm

Drones have become the go-to solution for various aerial tasks recently, but they haven't been around for that long.

What's surprising is that drones didn't require any groundbreaking new technology to emerge.

So, why did they take so long to become mainstream?

Comments URL: https://news.ycombinator.com/item?id=42139262

Points: 1

# Comments: 2

Categories: Hacker News

ChatGPT for Windows

Hacker News - Thu, 11/14/2024 - 1:12pm

Article URL: https://openai.com/chatgpt/desktop

Comments URL: https://news.ycombinator.com/item?id=42139234

Points: 1

# Comments: 0

Categories: Hacker News

GPT or BERT: why not both?

Hacker News - Thu, 11/14/2024 - 1:07pm

Article URL: https://arxiv.org/abs/2410.24159

Comments URL: https://news.ycombinator.com/item?id=42139161

Points: 2

# Comments: 0

Categories: Hacker News

Ask HN: Where do you go to learn data engineering patterns?

Hacker News - Thu, 11/14/2024 - 12:26pm

I am working on refactoring a system which I feel does some frequently encountered flows across organisations. Collects raw event streams, grabs unique URLs, crawls, extracts, creates vectors, uses these output in some downstream analysis and objects.

I have many different solutions which “work” but I am not sure which is optimal.

I find both the internet and LLMs to be a very poor resource for this type of work.

So where does one learn about how others have implemented such systems? Happy to reads books, blogs, repos!

Comments URL: https://news.ycombinator.com/item?id=42138619

Points: 1

# Comments: 0

Categories: Hacker News

The Poor UX/UI of Major Cloud Providers: ECS/VMs Management

Hacker News - Thu, 11/14/2024 - 12:26pm

The user experience of major cloud providers is surprisingly poor, especially when it comes to basic VM/instance management. Here are some major pain points I've encountered:

1. Complex Regional Segregation (AWS) - Instances are strictly segregated by regions - No unified view to manage instances across all regions by default (Need specific queries just to get a global view) - Sometimes you lose track of which region your instances are in

2. Instance Metadata Management is Unnecessarily Difficult - Can't easily rename instances (Google Cloud requires shutdown and restart) - No automatic tracking of which IAM account created the instance - Limited ability to add notes or tags for running services - When an instance runs multiple services (e.g., "service-A" actually running A, B, and C), there's no easy way to update its name or add proper documentation - Have to maintain external documentation to track what's running where - Need to SSH and check screen/tmux sessions to see what's actually running

3. SSH Key Management Issues (especially Google Cloud) - Can't properly manage SSH public keys - No ability to add descriptions to keys - Can only see a list of root public keys without context - No tracking of who added which key - No "last used" timestamp (unlike GitHub which handles this well)

I do use serverless solutions like Lambda/Cloud Functions and Firebase, but there are always cases where you need ECS/VMs. These basic UX issues make daily operations unnecessarily complicated.

What's your experience with cloud providers' UI/UX? Have you found any good solutions to these problems?

Comments URL: https://news.ycombinator.com/item?id=42138608

Points: 1

# Comments: 0

Categories: Hacker News

BoxLang

Hacker News - Thu, 11/14/2024 - 12:23pm

Article URL: https://www.boxlang.io/

Comments URL: https://news.ycombinator.com/item?id=42138572

Points: 1

# Comments: 0

Categories: Hacker News

Pages