Skip to main content

Hands-on agent demos. No sign-up, no sales call. Try Them Now

Book a Demo

R&D · Manufacturing · Pharma · Chemicals · Technology

Your Next Breakthrough Is Probably Already in Your Organization.

Somewhere in your lab notebooks, experiment reports, and research databases, a colleague has already done the work your team is about to start. Squirro finds it before the duplication happens. Trusted by Henkel across 3,300 researchers and 45 source systems worldwide.

Save Money Make Money Medium Effort Deploy in 4–8 weeks
Book a Demo Explore the Full Agent Catalog

Trusted by Industry Leaders

Zwickroell Henkel Siemens OCBC Bank Candriam Bertelsmann IDB Invest Security Benefit swissrockets Customer Logo Bühler logo

Measurable Impact

Colleagues working through requests together
$500K the cost of fragmented R&D intelligence
3,300+ researchers connected at our reference deployment for R&D knowledge discovery

Evidence in Action

Watch Squirro Uncover R&D Knowledge and Eliminate Duplicate Information End to End.

Loading the Squirro assistant…

How it works

Find Prior Work Before Launching the Next R&D Initiative.

The research exists. The problem is that it lives in silos across sites, systems, and the heads of scientists who may not be there tomorrow. Squirro connects every repository into a single knowledge graph that researchers can interrogate before starting a project to find out what has already been done, who did it, and what it found.

01

QUESTION

Researcher queries before starting a new project

Plain language search across the entire research corpus. No need to know which system holds what, or which colleague to email.

Plan Check Check a planned experiment against every experiment on record before a lab hour is spent · scored from indexed metadata Demo corpus · fictional company
Planned experimentLoad demo scenario
Objective
Scored against 30 experiments across 4 sites — materials, process steps and objective keywords, all indexed fields
MaterialsEpoxaCore 2100AluOx NanoFillCycloEp C11BioEpox B40SilOx FS-200
ProcessPlasma pre-treatmentSolvent-free formulationThermal curingRoll-to-rollSol-gel
Check for prior work
The same plan gives the same result every time — a budget decision can rest on it
02

CONNECTIONS

Every connected repository searched simultaneously

ELN systems, lab reports, SharePoint, patent databases, and external publications all queried in parallel. Knowledge graph surfaces non-obvious connections across sites, time periods, and research threads.

Research Assistant Every answer is grounded in the documents and the knowledge graph, with the tool steps on screen
Has anyone here worked on solvent-free epoxy barrier coatings?
Thought for 11.4s across 5 steps11.4s
Searched the archive — “solvent-free epoxy barrier coating PET” · 9 results1.2s
Loaded the knowledge-graph skill0.6s
Read the graph schema — Experiment · Material · Researcher · Site0.8s
Queried the graph — experiments usesMaterial EpoxaCore 2100 + AluOx NanoFill · 6 results2.1s
similar to · cross-site link derived by the graph — EXP-2024-117 ↔ EXP-2025-012 (Pune)0.9s
R&D knowledge graphQueried by the agent
EXP-2024-117→ EpoxaCore 2100→ AluOx NanoFill→ Dr. Wei Chen→ EXP-2024-021→ EXP-2025-012
29 of 108 nodes shown · experiments, materials, people, IP, sites · dashed edges are derived “similar to” links
The Pune retry surfaced through the graph — its report never uses the words in the question
03

DISCOVERY

Prior work, experts, and related patents surfaced

Relevant studies returned with full context. Internal experts identified. Related patents and publications linked. The researcher starts informed, not from scratch.

Plan Check Top 5 related prior experiments · scored from indexed metadata Ask the archive
Likely duplicate: EXP-2024-117 FAILED
Failed (thermal degradation) at Suzhou. Estimated avoided spend: $340,000 · 28 weeks — the recorded cost and duration of EXP-2024-117, in this demo.
ExperimentOutcomeSimilarity
EXP-2024-117 Solvent-free epoxy barrier coating on PET film with plasma pre-treatmentSuzhou · led by Dr. Wei Chen · cure below 110 °C or switch to cycloaliphatic epoxy FAILED88% similar
EXP-2024-021 Plasma-assisted barrier coating on BOPP filmSuzhou · $175,000 · 13 weeks on record SUCCESS61% similar
EXP-2025-012 Bio-epoxy barrier coating, plasma-assisted (Phase 2)Pune · running now · BioEpox B40, cured below 110 °C ONGOING50% similar
Expert Finder · Barrier coatings#1 Dr. Wei Chen · Suzhou
Principal Scientist, Barrier Materials · score 18 · ranked from a pending patent, two successful barrier studies, a publication — and EXP-2024-117, the failure this plan collided with
Knows why it failedRequest collaborationdelivery simulated
Every number on this screen resolves to a record you can open
04

FLYWHEEL

Organizational knowledge compounds over time

Every completed study enriches the knowledge graph. Every connection made strengthens the network. The organization gets smarter with every experiment, whether or not the scientist who ran it is still there.

R&D Landscape What the archive holds and what its failures cost · every figure is an aggregate over indexed records counted from the archive
$1,495,000Failed-experiment cost on record · 8 experiments
$340,000 · 28 wkLargest avoidable repeat · EXP-2024-117
30Experiments on record · 8 failed · 3 ongoing
18 · 4Researchers · sites
8 · 8Patents · publications linked
16Materials tracked
Failed-experiment cost by failure mode$1,495,000 in this demo archive
Scale-up instability$480,000
Thermal degradation$340,000
Phase separation$280,000
VOC non-compliance$180,000
Delamination · Adhesion loss$215,000
Failure mode is an indexed field on every failed-experiment report — this breakdown can run as a scheduled report

Integrations

Connect Seamlessly with Your Existing Data

Break down data silos and integrate all of your existing data, creating a single source of truth.

Data source connectors for the R&D Knowledge Discovery & Deduplication Agent

Your Expansion Path

Start Here. Scale Across Innovation Intelligence.

Each stage shortens discovery cycles, reduces duplication costs, and builds a compounding knowledge advantage across the organization.

  1. 01 To start

    R&D knowledge discovery & deduplication agent

    Semantic search across all research repositories, knowledge graph linking experiments and experts, prior work surfaced before new projects begin. Live in 4–8 weeks.

  2. 02 Next

    Innovation intelligence agent

    AI-assisted monitoring of external research, competitor patents, and scientific literature surfaced alongside internal knowledge for a complete picture of the innovation landscape.

  3. 03 Next

    Patent landscape monitoring agent

    Internal IP connected to external patent databases. Freedom-to-operate insights surfaced automatically before research investment is committed.

  4. 04 Vision

    AI-assisted product development

    A connected R&D intelligence layer spanning internal knowledge, external literature, patent landscape, and expert networks that compounds with every study, accelerating every discovery cycle.

Our Platform

Connect Once, Govern Once, Reuse Everywhere.

Most enterprises rebuild the same plumbing for every AI project — connectors, governance, retrieval, audit. Squirro builds it once. You extend it.

Squirro Platform Copy-1

One Platform. Limitless Workflows. Ready to Deploy.

Ready to Deploy

Start Where the Friction Is.

R&D knowledge search, innovation intel gathering, and patent monitoring each create friction differently in every org. We work with R&D teams to map the most strategic path to value and to deploy quickly where the friction is sharpest.

Share your details with us to discover the use case, discuss how it fits your stack, or explore a deployment.

Browse the Catalog