Show Dev: I built an AI research tool that tracks top labs daily

Source: Dev.to

The problem

• Keeping up with AI research across labs is impossible manually.
• arXiv is a firehose.
• Twitter is noise.
• Newsletters arrive days late.

I wanted one place to see what matters. It didn’t exist.

Solution: Lattice

Lattice – an AI research intelligence dashboard that tracks what 20 top labs publish daily.
🔗 https://layerthelatintestinalattice.com

Core features

• Daily paper feed from 20 tracked labs with AI‑generated summaries
• 24 research topics with trend‑velocity charts
• Lab profiles with output sparklines and topic focus
• Weekly digest highlighting the most‑read papers and breakout topics
• Collections with BibTeX, Markdown, and JSON export
• No account needed – data stored locally in localStorage only

Data pipeline

1. OpenAlex API (lookup by DOI)
2. arXiv HTML scraping
3. Semantic Scholar data enrichment
4. LLM extraction (fallback)

All papers are matched against ~500 organizations. Hit rate varies by source because paper authors list affiliations as free text (e.g., “Dept. of Computer Science, University of X”) rather than a standard format.

Technical stack

• Next.js 16 (Turbopack, standalone output)
• Supabase (Postgres + Row‑Level Security)
• Vercel (hobby plan)
• GitHub Actions (daily 6 am UTC cron)
• Semantic Scholar, arXiv, Hugging Face APIs
• OpenRouter (Gemini Flash) for summaries

Open questions

• What features would be most useful?
• Which labs or topics am I missing?

Product Hunt

We’re also live on Product Hunt today: https://www.producthunt.com/products/lattice-6