Company research
Data as of June 10, 2026
No company research is published for this company yet. All research →
Unstructured
ETL for unstructured data preprocessing
AI InfrastructureFounded 2022#16 of 55 in AI Infrastructure
Updated June 10, 2026
Follow this company
Follow this company to revisit its latest research cards from your account.
Company profile
What is Unstructured?
Unstructured specializes in extracting and transforming complex unstructured data from diverse sources like PDFs, images, and emails into clean, structured formats. Their tools are critical for preparing high-quality input data for large language models (LLMs) and other AI applications, enabling organizations to build more accurate and effective AI solutions.
History
Research history
8 research updates
History
Research history
Ai Infrastructure
Unstructured: GitHub commits decreased 100% over the last 30 days
LatestAi Infrastructure
Unstructured: GitHub commits decreased 100% over the last 30 days
2 metricsAi Infrastructure
Unstructured: GitHub commits decreased 100% over the last 30 days
2 metricsAi Infrastructure
Unstructured: GitHub commits decreased 100% over the last 30 days
2 metricsAi Infrastructure
Unstructured: GitHub commits decreased 96.7% over the last 30 days
2 metricsAi Infrastructure
Unstructured: GitHub commits decreased 97.4% over the last 30 days
2 metricsAi Infrastructure
Unstructured: GitHub commits decreased 97.7% over the last 30 days
2 metricsAi Infrastructure
Unstructured: GitHub commits decreased 98% over the last 30 days
2 metricsLatest company data
Primary data point
Package downloads (30d)Metric: PyPI downloadsSource: PyPI BigQueryScope: UTC activity dayUpdates: DailyFreshness: 48h target; 48h expected source lagConfidence: HighCaveat: Includes all package installations including CI/CD pipelines and mirrors; pypistats is fallback only.Methodology →
5.6M/30d
Tracked package: PyPI unstructured
▲ +15%Updated 1d ago
Other data points
Dependent projectsMetric: PyPI dependentsSource: Libraries.io APIScope: UTC daily snapshotUpdates: DailyFreshness: 48h target; 48h expected source lagConfidence: HighCaveat: Counts direct reverse dependencies from PyPI.Methodology →
295
Projects depending on tracked package: PyPI unstructured
Updated 15h ago
GitHub starsMetric: GitHub starsSource: GitHub GraphQL + GH ArchiveScope: UTC daily snapshotUpdates: DailyFreshness: 48h target; 24h expected source lagConfidence: HighCaveat: Stars are bookmarks, not usage.Methodology →
14.9K
Main repository stars
Updated 15h ago
Hacker NewsMetric: Hacker News mentionsSource: Hacker News Algolia APIScope: UTC activity dayUpdates: DailyFreshness: 48h target; 24h expected source lagConfidence: HighCaveat: Search relevance and historical availability come from Algolia's HN index.Methodology →
63/30d
Position #9 in category discussion
Updated 15h ago
Repository health
Maintenance data from the main open-source repository.
Key-person riskSource: GitHub public event dataUpdates: WeeklyCaveat: Minimum developers needed for half of recent code changes. Lower = higher concentration risk.Methodology →
2
External contributorsSource: GitHub public event dataUpdates: WeeklyCaveat: Based on push events; may undercount contributors who only open PRsMethodology →
40
% of recent contributors outside the core team
Releases (30d)Source: GitHub release dataUpdates: WeeklyCaveat: Published releases in the last 30 daysMethodology →
0
Repository usage
Public repositories and source files importing packages tied to Unstructured.
Repos importingMetric: Public GitHub repos using company packagesSource: ecosyste.ms Packages APIScope: UTC daily snapshotUpdates: DailyFreshness: 48h target; 48h expected source lagConfidence: HighCaveat: Public GitHub dependency data only; private repos and direct API use outside those packages are not visible.Methodology →
3.4K0%
About Unstructured
Unstructured specializes in extracting and transforming complex unstructured data from diverse sources like PDFs, images, and emails into clean, structured formats. Their tools are critical for preparing high-quality input data for large language models (LLMs) and other AI applications, enabling organizations to build more accurate and effective AI solutions.
FoundersBrian Raymond, Matthew Harrison
Funding
$105M · 3 roundsRaised $105M total. Category position #16 of 55 in AI Infrastructure.
Series B2024
$40MMenlo Ventures
Series A2024
$40MMenlo Ventures
Seed2023
$25MMadrona
Investors
MadronaMenlo Ventures
Tracked packages (1)
1 PyPI
unstructured
PyPIMain PyPI packageunstructured
ETL for unstructured data preprocessing