HN Showcase
Back to browse
What is HN thinking? Real-time sentiment and concept analysis

What is HN thinking? Real-time sentiment and concept analysis

by ddtaylor·Feb 12, 2026·37 points·23 comments
Visit ProjectView on HN

AI Analysis

●●SolidDark HorseSolve My Problem

HN sentiment tracker for under $1/month using llama-3.1; niche but genuinely useful.

Strengths
  • •Exceptional cost efficiency: achieved full pipeline (extraction, embedding, storage, dashboard) for <$1 infra spend
  • •Real-time concept and entity tracking provides novel lens on HN discourse that platform doesn't natively expose
  • •Multi-dimensional analysis (sentiment, controversy, depth, discourse type) offers richer signal than simple voting
Weaknesses
  • •Narrow audience: meaningful mainly to HN meta-watchers and discourse researchers, not general users
  • •Architecture trades off LLM/embedding family coherence (llama-3.1 + higher-dim embedding) for cost; unclear if analyses remain coherent
Category
Data
Target Audience

HN community analysts, data enthusiasts, researchers studying online discourse

Similar To

HN API clients · News aggregators with NLP · Discourse analysis platforms

Post Description

Hi HN,

I made Ethos, an open-source tool to visualize the discourse on Hacker News. It extracts entities, tracks sentiment, and groups discussions by concept.

Check it out: https://ethos.devrupt.io

This was a "budget build" experiment. I managed to ship it for under $1 in infra costs. Originally I was using `qwen3-8b` for the LLM and `qwen3-embedding-8b` for the embedding, but I ran into some capacity issues with that model and decided to use `llama-3.1-8b-instruct` to stay within a similar budget while having higher throughput.

What LLM or embedding would you have used within the same price range? It would need to be a model that supports structured output.

How bad do you think it is that `llama-3.1` is being used and then a higher dimension embedding? I originally wanted to keep the LLM and embedding within the same family, but I'm not sure if there is munch point in that.

Repo: https://github.com/devrupt-io/ethos

I'm looking for feedback on which metrics (sentiment vs. concepts) you find most interesting! PRs welcome!

Similar Projects

Data●●Solid

Opportunities from Hacker News Comments

The site turns HN into an idea-finding feed — tabs for Most Positive/Most Negative and an 'Opportunities' view plus per-story 'View Analysis' make it immediately usable. It's a tidy UX for skimming pain points, but the landing page gives no detail about the extraction accuracy, deduplication, or ranking method, so the real value depends on the unseen NLP and heuristics.

Niche GemSolve My Problem
hnsignal
103mo ago