Structured summaries
Each paper is distilled into a concise summary highlighting the information our AI judges most relevant - including findings, context, and limitations. The original paper is always one tap away.
Research tool for scientists & clinicians
Stay on top of your field in 60 seconds a day.
SciEspresso delivers personalized summaries of newly published peer-reviewed research - matched to your field, searchable by meaning, and available as daily audio briefings. Built by researchers.
Free to start. No account needed. Built by researchers.
Early restricted fluid resuscitation versus usual care in septic shock
RCT
n = 1,563
24 centres
Complexity 6/9
Key finding
Restricting fluids during the first six hours did not reduce 90-day mortality (24.1% vs 25.4%; 95% CI −4.8 to 2.2). Vasopressor-free days and renal replacement therapy were unchanged.
Similar →
Today's briefing · Internal Medicine
Transcript excerpt
"First up: a large multicentre trial on fluid resuscitation in septic shock finds no mortality benefit to early restriction. Next: a new meta-analysis links short sleep to next-day working memory deficits. And three papers on GLP-1 agonists worth your attention this week..."
Audio archive →
How it works
Three steps, no setup.
Open the app, describe your research interest, and start reading.
01
Search the archive
Type a clinical question. Search covers every indexed paper, from this morning’s publications back through the archive.
02
Read the structured summary
Study design, sample size, setting, and the key finding - every summary links to the original paper.
03
Explore similar papers
Jump from any summary to a list of similar papers, matched by meaning rather than shared keywords.
What you get
Everything a researcher needs in one place.
Similar papers
Every paper comes with a list of similar papers, matched by meaning rather than shared keywords, so you can see related work at a glance.
Standing Searches
Describe your research interest and get a nightly paper scan matched to your field. Matching papers appear in your feed with a personal badge.
Archive and audio
A searchable archive of indexed papers, plus weekly audio briefings by research field, perfect for commutes and lab time.
Coverage
Many research areas. One daily briefing.
From clinical medicine and neuroscience to AI, climate, and beyond - all in one place.
Health & Medicine
"Early restricted fluid resuscitation versus usual care in septic shock"
Mental Health & Sleep
"Short sleep duration and next-day working memory: a meta-analysis"
Tech & AI
"Large language models for triage support in emergency departments"
Climate & Environment
"Urban heat exposure and cardiovascular admissions across 40 cities"
Aging & Longevity
Brain & Behaviour
Women's & Men's Health
Fitness & Nutrition
Parenting & Child Development
Food & Agriculture
Consumer Tech & Wearables
Energy & Sustainability
Space & Earth Science
Society & Economics
Work & Productivity
Pets & Animals
All coverage areas →
The app
iOS and Android.
Free to download on both platforms.
Pricing
Free to search. €1.99 to stop counting.
Search across the whole archive is free on both plans. The subscription adds Standing Searches and removes the caps.
Single Shot
FreeFor researchers getting started
- 50 summaries per week
- Search across the full archive
- Similar papers on every summary
- Papers from the last 14 days
- Standing Searches
- Audio briefings
Double Shot
€1.99 / monthFor active researchers and clinicians
- Unlimited summaries
- Search across the full archive
- Similar papers on every summary
- Up to five Standing Searches, with a notification when new papers land
- The complete archive
- Unlimited audio briefs
Billed in-app · cancel anytime
Who builds it
Built by researchers, for researchers.
SciEspresso is made by a small team of researchers and clinicians based in Munich. We built the tool we wished we had during our own research careers - precise, fast, and honest about what AI can and cannot do.
What we will not do
Summaries never replace the paper. Every one links to the original source, states the study design it came from, and carries a limitations note. AI-generated text can be wrong, and we say so where it matters rather than in the small print.