Nicheloom visits: 58,213 all-time · 7,637 in the last 7 days

The opportunity tracker for new startups.

Today: 1,008 launches · Niche of the day: specialized calculation tools · See what launched today →

Athenic

Why you can't do data analysis with Claude

Get picks like this daily. Today's launch numbers, the top launches, AI and tech news, and a daily opportunity pick, straight to your inbox.

Want to reach these readers? Sponsor an issue.

Hey HN, I'm Jared. I’ve been building data tools since 2020. Polyture, then AskEdith, now Athenic: ask a question in natural language, get a chart/dashboard, then automate it. Connects to Postgres, Salesforce, Google Ads, whatever.To everyone that says "just link Claude to your db”: imagine the chaos of conflicting definitions and analysis that would show up in a business setting.Ask “what’s our revenue?” twice, two days apart or to a different model. There’s no guarantee that you’ll get the same results. Now imagine giving that to all of the non-technical users at your company.It's not a model problem. We learned this the hard way. When we launched AskEdith here in 2022, and you told us (https://news.ycombinator.com/item?id=33435361): “you'll still have to check the SQL,” “trust is everything,” “answers won’t be consistent”. You were right.Now, Athenic defines KPIs and formulas deterministically in a Semantic Model. The Semantic Model is made up of modular, composable units that can make up complex analysis, while guaranteeing determinism and accuracy. The LLM’s only responsibility is interpreting your question (which even non-technical users can double check).`revenue = sum(order_total − refunds) where status = 'completed'`Ask for revenue and everyone gets the same number, every time.Three years of learnings from working with top startups and Fortune 500 companies later, we just shipped 2.0. Chat-to-insight, plus dashboards and automations that run on a schedule and land in your email. Tell us we're wrong.

HN · launched Jun 10, 2026 · 5 upvotes (top 87%) · 0 views

Niche: ai intelligence and analysis tools · 141 launches, 17 in the last 7 days ▲ +13% · Explore this niche →

Category and tags

  • Horizontal
  • Analytics & BI
  • B2B
  • AI-native
  • Hobby / open-source project

Visit website ↗ Source listing ↗

Could you build this?

No Building an AI data analyst that reliably translates natural language to complex multi-table SQL across diverse schema types without hallucinations requires deep data engineering, semantic layer modeling, and robust evaluation infrastructure.

What it would actually take: A production version requires connector infrastructure (JDBC/ODBC pooling for Postgres, Snowflake, BigQuery, third-party APIs), a semantic metadata engine that generates and verifies join paths, an agentic loop with self-correcting SQL execution against sandboxed read replicas, and deterministic charting synthesis. The hard part is high precision on ambiguity and avoiding harmful or hallucinatory queries across large enterprise schemas.

Discussion

No comments on this launch.

Competitors

Other products that read as similar to this one : 147 launches clear the similarity bar, closest 8 shown.

Attention rank: #132 of 148 (itself plus its competitors, highest first, normalized so YC and Product Hunt are compared fairly).

Launched 219 days after the earliest competitor.

The niche, labels, vibe code rating and competitor comparison on this page are analysis by Nicheloom. If you use them, please credit nicheloom.com. How you may use our data.