Published 26 May 2026· via Google Cloud· Customer story
Junson Capital automates global investment reporting with Vertex AI, Gemini and Document AI
Junson Capital · Finance · Hong Kong
The source does not publish a single comparable figure. Reported results are listed below.
Summary
Junson Capital is a global investment management firm that built an AI-native data engine on Google Cloud to extract structured intelligence from high-variance, unstructured documents.
The firm uses Vertex AI, Document AI, and Gemini models to process private bank statements, inspection testing reports, and handwritten documents with a proprietary six-layer recognition framework.
The implementation also supports regional hosting for data residency and powers an AI-native digital employee workflow for operational tasks.
The problem
- 01Extract structured intelligence from high-variance, unstructured documents.
- 02Reduce the 15.5 days required to generate a single report.
- 03Automate manual internal reimbursement and invoice matching workflows.
What they built
- 01Built a proprietary six-layer recognition framework using Vertex AI and Document AI for document extraction.
- 02Used Gemini 2.5 Flash and Gemini 2.5 Pro for document understanding and hosted Gemini in Google Cloud regions for residency.
- 03Created an AI-native 'digital employee' workflow that integrates extracted data into operational processes.
Reported results
- 01Reduced report generation time from 15.5 days to 1 day.
- 02Reduced monthly internal reimbursement workflows from 10 days to 1 hour.
- 03Achieved 90%+ resource efficiency in data extraction.
- 04Enabled 2 people to handle work that previously required 15.
- Built AI-native document extraction engine
- Automated report generation workflow
- Automated reimbursement and invoice workflows
- Enabled regional data residency hosting
Source
How this deployment compares
Computed from the library · data 10 Nov 2026Junson Capital has 3 deployment records in the library. It is one of 71 intelligent document processing deployments, 12 of them in Finance. Like 63% of records, its source gives no figure.
Junson Capital in the library
3 deployment records from Junson Capital
- May 2026This record · Intelligent document processing
- Jun 2026Junson Capital scales an AI-native data engine with Vertex AI and Document AI (investment reporting + reimbursement automation)Intelligent document processing
- UndatedJunson Capital: AI-Native Investment Reporting with Vertex AI and Gemini on Google CloudDocument automation
Too recent to judgeFollow-through is judged once a record is 1 year old; this one is 6 months old.
Same pattern, whole library
One of 71 intelligent document processing deployments
- Finance accounts for 12 of them (17%), ranking #2 of 12 industries.
- 50 records of this pattern came before it; 4 have been published since.
- Only 13 of the 71 put a number on the result (18%).
- AWS55%
- Microsoft25%
- GCP20%
- Amazon Textract 28
- Amazon S3 18
- Amazon Bedrock 18
- AWS Lambda 11
- Amazon Comprehend 8
- Power Automate 7
Outlined: used in this record.
Shares describe this library, not the whole market.
Results against peers
No figure here. What peers report
Like 63% of records in the library, the source gives no figure. Peers that do quantify similar claims report:
- Time & speed36 figures, Finance · middle half 39%–75%57%median reduction
- Risk, reliability & safety13 figures, all industries · middle half 60%–90%70%median reduction
Peer figures are as reported by their sources; percentage claims only.
How we checked this record
Evidence tier: supported
A credible source with a named customer, but not the organization's own account.
- Named organization
- Junson Capital
- Who published the source
- The organization itself
- Deployment status
- Production (confidence 95%)
- Quantified result
- The source reports no figure
- Summary checked against source
- 4 / 5 agreement
Last re-checked 13 Jun 2026.
This record is evidence for
Same pattern, other organizations
All intelligent document processing deployments- 8 Jun 2026HealthcarePR
Paynela automates healthcare payment adjudication with generative AI using Amazon Textract and Amazon Bedrock
Paynela · Intelligent document processing · AWS
Paynela is a Puerto Rico-based healthcare financing company that helps patients manage out-of-pocket medical expenses and patient assistance program workflows. The company had a manual medical adjudication process that created backlog and delayed patient payments. Paynela combined OCR with Amazon Textract and Amazon Bedrock to analyze complex healthcare data and automate payment adjudications.
- 6 Jun 2026OtherItaly
DeAgostini automated 54,000 supplier invoices using SAP Document AI
DeAgostini · Intelligent document processing · GCP
DeAgostini used SAP Document AI to automate invoice handling for supplier invoices. The solution focused on processing paper- and PDF-based invoices without disrupting operations, reducing manual effort and errors.
- 3 Jun 2026FinanceHong Kong
Junson Capital scales an AI-native data engine with Vertex AI and Document AI (investment reporting + reimbursement automation)
Junson Capital · Intelligent document processing · GCP
Junson Capital is a global investment management firm spanning real estate, private equity, and public markets across Hong Kong, Singapore, New York, and Palo Alto. The firm built a proprietary AI-native data framework to extract structured intelligence from high-variance, unstructured investment and administrative documents. The solution uses Vertex AI, Document AI, Gemini 2.5 Flash, Gemini 2.5 Pro, and Google Cloud Regions to support data extraction, automation, and regional data residency.
Production
- 2 Jun 2026HealthcareUnited States
Automating contract intelligence with Doczy.ai on AWS
AArete · Intelligent document processing · AWS
AArete’s Doczy.ai automates contract intelligence for healthcare organizations and other enterprises by converting unstructured contracts and legal documents into structured, queryable outputs on AWS.
Agentic