PatentsUnwatched
3.7M / yr
The filing that blocks your prosecution path was submitted 18 months ago. You'll find out in due diligence.
Lapis · Live feed
72,000 patents are filed every week. Your team can’t read all of them.
Lapis deploys autonomous researchers that surface threats, opportunities, and blind spots across the patents, research papers, and competitor news in your field.
Lapis · Live feedNo query required
Lapis · Live feed--:--:-- UTC
New research surfaced since you opened this page
0.00[01]
6,000
papers / day
[02]
72,000
patents / week
[03]
<1%
any team reads
Sys // Perception-01
Sample output. Production scouts run against your private roadmap, IP, and watchlist.
Backed by · Built with · Featured in
Founders Hub
Research Partner
Toronto Network
Schulich Network
Featured · 2025
Founders HubResearch PartnerToronto NetworkSchulich NetworkFeatured · 2025Start here
See it working in your field in three steps.
Step 01 · Tell us your space
Pick the field you work in.
Lapis prefills your scouts with sources, ontologies, and the right watchlist for your industry.
You’re making million-dollar decisions on a fraction of the available science.
Most teams find out too late. That’s not bad luck. It’s a volume problem.
Patents publish 18 months after filing. The science behind them was happening years before that. 6,000 papers a day, 72,000 patents a week. No team on earth is reading even one percent of it.
New research papers published since you opened this page
0.00
Volume vs. attention
Sources: NSF Science & Engineering Indicators · ScopusMove across the curve · 1996 → 2024
PapersUnwatched
3.5M / yr
The lab that's 14 months ahead of your program published in March. Nobody on your team saw it.
GrantsUnwatched
$92B / yr
The NIH just funded your competitor's next two years of research. It's public record. You didn't know.
RegulatoryUnwatched
1.2M+ / yr
A rival's filing just narrowed the approval path your program depends on. It sat in a public docket for months.
CompetitorsUnwatched
70K+
The startup that just announced in your space wasn't stealth. They were just publishing in places you weren't watching.
PreprintsUnwatched
600K+ / yr
The result that reshapes your roadmap won't be peer-reviewed for a year. It's already posted. The race started without you.
02 / How it works
Lapis runs a team of scouts on your field.
A scout is a research agent that reads one kind of source, all day, and never gets tired. Together they learn your program, read the new science, rank what is relevant to you, and send a short brief.
Learn your program
Scouts learn what you're working on, in private.
Share your roadmap, IP portfolio, and watchlist. Scouts privately index it and learn what counts as real news for your team, and what doesn't.
Book a call ↗The most powerful part: they work for you every day, even when you’re not looking. While your BD lead is on a plane and your CEO is in board meetings, Lapis is reading the 6,000 papers that dropped overnight.
The bottleneck moved
Not because it’s hidden. Because no one has time to read it. Lapis does.
03 / The Field
YOUR WORLD, CHANGING
Distance is relevance to your program. Brightness is intensity. Dashed rings are uncertainty. Watch the quiet — then watch it break.
LNP-7 · HEPATIC-TROPIC mRNA DELIVERY
Your therapeutics program — your program
COREADJACENTPERIPHERYHORIZON
QUIET
3 scouts active · 0 documents read
Hover a point to learn why it matters to you · click to open · the field replays on demand
How it perceives
Four principles behind every scout.
WEAK SIGNALS
Scouts read the things you did not know to search for. A signal can stay quiet for weeks — Lapis keeps it, until it starts to matter.
SUDDEN CONVERGENCE
Several unrelated findings can resolve into one larger pattern. The real story is rarely a single document — it's the movement behind it.
VISIBLE UNCERTAINTY
Every claim carries its confidence. A weak signal looks weak. Lapis never hides how sure it is.
WHERE ATTENTION MOVES
It shows not only what exists, but where you should look next — and why it matters to your program, not to the field in general.
Live demo
The whole landscape, on one canvas.
Pan across the terrain, zoom into an active field, and open any summit or white-space basin. Elevation reflects live research and filing density.
Booting console…
Drag to pan · scroll to zoom · click a summit or basin
Global coverage
Global science, watched continuously.
Every hub, filing, and collaboration — tracked across the labs and companies moving your field. Grab the globe and spin it.
Loading country boundaries · drag to rotate · select a city
─── What you get
Here’s what that looks like.
Not a list of links. A brief, with reasoning, sources, and a clear next step. The output of an entire research team that ran through the night, distilled to what actually matters for your program.
Lapis · BriefDelivered 2d ago
High signal · Competitive IP · 2026-W21
Helix Therapeutics is 14 months ahead on hepatic-tropic LNP delivery.
From Atlas IP / Patents · Mira Literature · Felix Grants
A continuation filed last Thursday (USPTO 18/421,887) overlaps claims 3–7 of your LNP-7 prosecution path. Cross-referenced against a March bioRxiv preprint and a $1.8M NIH R01 from the same lab. This isn’t a one-off filing, it’s a coordinated program.
Convergence signal · 3 sources, one lab
Preprint · bioRxiv · 2026-03-04
“Branched ionizable lipids enable selective hepatocyte delivery of large mRNA payloads”
Schaefer Lab, TU Munich · 4 authors · 87 citations
Grant · NIH R01 · awarded 2025-09
GM-148312 · $1.8M over 4 years · same lead investigator
Mechanism: hepatic LNP delivery for mRNA therapeutics
Patent · USPTO continuation · 2026-05-15
Application 18/421,887, claims 3–7 overlap LNP-7 chemistry
Two prior-art references appear missing from prosecution
Why it matters
If granted as filed, claims 3–7 would constrain your hepatic delivery program in US and EU markets, directly affecting your Q3 prosecution plan.
Recommended next step
Review claims 3–7 against your prosecution strategy. We’ve flagged two prior-art references they may have missed, both linked in the full brief.
3 sources cited · full reasoning in appOpen full brief
695M+ IP records · monitored live
Patent offices. Research databases. Industry signals.
Patents, papers, and industry signals, ingested the moment they publish. The set below is a live sample; we bring new feeds and jurisdictions online every week.
USPTO
+7K/week
EPO
+2.2K/week
WIPO
+5.5K/week
KIPO
+4K/week
JPO
+6K/week
IP Australia
+550/week
CIPO
+495/week
arXiv
+6.5K/week
PubMed
+30K/week
Semantic Scholar
+50K/week
OpenAlex
+125K/week
bioRxiv
+1.2K/week
CORE
+65K/week
Europe PMC
+22K/week
Hacker News
+40K/week
Product Hunt
+300/week
TechCrunch
+200/week
Y Combinator
+12/week
Google News
+50K/week
+ many more being added
695M+ records indexedPatents · papers · competitors
─── Why not just use ChatGPT
ChatGPT is great if you know what to ask. The hard part is knowing what you don’t know.
Once you know about a specific patent, anyone can summarize it. Finding it in the first place, across millions of documents, in 18-month-old data, before it shows up anywhere obvious, is the job Lapis does.
LapisAnswers questions you already know to ask
Surfaces the patent, lab, or grant you didn't know existed
Reads what you paste in
Reads every filing and preprint in your field, end to end
Stops when you close the tab
Runs every day, on a schedule, while you're doing your job
Generic answers, hallucinated citations
Cited findings from primary sources, with reasoning
─── Try Lapis free
Get a free sample brief.
Pick one program. We’ll send you the brief within a week.