Live now · Hacker News → Linear

Pain signal,
on tap.

An agent watches Hacker News for what your customers are actually saying. It classifies what matters. It files it in Linear. Twenty-four hours from sign-up to your first issue.

Hacker News
NVSBL classifier
Linear
The problem

Your customers are telling you what to build.

They're doing it on Hacker News right now — in a Show HN comment thread, in a side-bar gripe under a launch post, in an Ask HN nobody on your team has the time to skim. Most product teams know HN is full of signal and ignore it anyway, because the math doesn't work: too much volume, not enough hours.

NVSBL does the reading so your team doesn't have to. We watch HN around the clock, classify what's actually pain versus what's chatter, and file the signal as Linear issues your team can triage like any other ticket. No new dashboard to log into. No browser tab to leave open. No daily ritual to forget.

What you get

Three jobs. One pipeline.

Hacker News, watched for you

Front page, Show HN, Ask HN, every comment thread. Filter by score threshold, by keyword, or by a one-line description of what your category looks like.

Agentic classification

Not keyword matching. We classify each signal by category — feature request, bug, competitor mention, churn risk, raw pain, opportunity — using a model tuned to your domain. You get signal, not search results.

Filed straight into Linear

Each signal becomes a Linear issue, in the team and project of your choice, with the original quote, the source link, the classification, and a confidence score. Triage it like any other ticket.

Sources

Today, Hacker News.
Tomorrow, you tell us.

Live now

Hacker News

Front page, Show HN, Ask HN, Tell HN, every comment thread. Score-threshold filters, keyword filters, and a one-line 'category description' you give us that becomes the semantic match.

On the roadmap — vote what's next

RedditLinkedInX / TwitterProduct HuntDiscordApp Store reviewsG2CapterraTrustPilotSubstackIndie HackersStack Overflow
See the roadmap →
Pipelines

Today, Linear.
Tomorrow, you tell us.

Live now

Linear

Every classified signal becomes a Linear issue, in the team and project you choose. Title generated from the signal; description includes the original quote, the HN source link, the classification label, and a confidence score. Filter by classification: send all of it, or only feature requests, only bugs, only competitor mentions.

On the roadmap — vote what's next

SlackNotionAsanaJiraGitHub IssuesClickUpHeightWebhooksEmail digestDiscord webhook
See the roadmap →

Custom destination not on the roadmap?

Just ask. We deliver. Most custom pipelines ship within two weeks at a flat $1,500 per build. Scale customers get one per quarter at no charge.

Request a pipeline →
How it works

Set it up once. Forget it exists.

01

Sign up

Pick a plan, give us a card. Two minutes.

02

Tell us what to listen for

One sentence describing your category, plus any HN-specific filters you want. We provide starting templates by category.

03

Connect Linear

OAuth into Linear, pick the team and project, decide which classifications get filed.

04

Open Linear tomorrow

Your first signal will already be there. Triage it. Correct anything the classifier got wrong. The system learns fast.

Pricing

Honest pricing.

Three plans, scaled by Linear-issue volume. Every plan runs the live pipeline. Cancel anytime.

Starter

$29/month

or $278/year

Up to 200 classified Linear issues per month. Single Linear workspace. Email support. Fourteen-day free trial, no card required.

Start with Starter →
Most popular

Team

$99/month

or $950/year

Up to 1,500 classified Linear issues per month. Multiple Linear teams and projects. Priority queue. Slack support.

Start with Team →

Scale

$399/month

or $3,830/year

Unlimited signals. Custom classification model. Bring your own model / API key. One custom pipeline build per quarter. Direct Slack channel.

Talk to us →
FAQ

Answers, briefly.

Is this just social listening?

No. Social listening tools watch for brand mentions. We watch for pain — feature requests, friction, churn signals, competitor complaints — and classify it so it lands in the right shape in Linear. The output is a triageable issue, not a dashboard you have to remember to open.

Why only Hacker News at launch?

We'd rather ship one source extremely well than four sources poorly. HN was first because it's the densest signal-per-post platform on the open web for our category — and because it's the place our own customers already read. The next sources are user-voted; the roadmap is public.

Will it spam Linear?

Only if you ask it to. By default, you set a daily issue cap and a classification filter. Most teams start with 'feature requests only, capped at ten per day' and tune from there.

Where does the data come from?

Public Hacker News content only — the official HN API and its public archive. No scraping, no private content, no logged-in data.

Do you train models on our setup?

No. Your filters, your classifier corrections, and your usage are not used to train any global model. Each customer's classifier is tuned against their own examples in their own workspace.

Can I bring my own model / API key?

Yes, on the Scale plan. BYO OpenAI, Anthropic, or self-hosted endpoint.

How do I influence what ships next?

Every customer gets a vote on the public roadmap. Sources, destinations, and features are all up for vote. The most-voted item ships first. We tell you when your votes ship.

The best designs are NVSBL.

Set it up once. Let it run. Stop scrolling Hacker News for signal — start receiving it in Linear.