iTromsø · IBM, VC2
10 minutes per day — time spent reviewing potential stories
Independently reportedMedia & entertainment · Document processing · NLP · In production
Document processing
Department of Veterans Affairs · Aquia Nava II LLC
Vendor-reportedPublic sectorUnited StatesScaledPublic sectorNLP
Vendor-reported. The customer is named and the numbers are quoted from the source page, but the account comes from the vendor. No independent confirmation.
Reported by aquia.us (vendor self-report). Checked against the source page on 2026-09-05. 4 of 4 figures below appear on that page word for word.
The problem
Approximately 25% of disability benefit claims couldn't be automatically classified, requiring manual review by VSRs and delaying decisions for Veterans.
What was deployed
- Vendor
- Aquia Nava II LLC
- Products
- Python FastAPI, Amazon Web Services (AWS) S3, Amazon Elastic Container Registry (AWS ECR), Datadog
- Technique
- NLP
- Build or buy
- Bought and customised
- Deployment
- Hybrid
- Data used
- Veterans' descriptions of medical conditions
- Scale
- thousands of claims daily
What changed
Each row is quoted from the source. Figures we could not find on the page in those words are marked — they are kept, not deleted, so you can judge them.
100%Conditions automatically classified upon submissionfrom 74%
“100% of conditions are now automatically classified upon submission — up from just 74% before the expanded classifier went live.”
Quoted word for word from the source
nearly 20,000Additional conditions automatically categorizedeach week
“nearly 20,000 additional conditions are automatically categorized each week”
Quoted word for word from the source
Manual classification work
“The automation has reduced manual classification work by approximately 80%”
Quoted word for word from the source · round percentage, no baseline given
4,131VSR working time savedannually
“saving VSRs an estimated 4,131 hours — or 172 days — of working time annually”
Quoted word for word from the source
sub-secondProcessing times
“sub-second processing times”
Quoted word for word from the source
Difficulties and limits
- Medical terminology varies widely and evolves constantly
- Veterans describe the same condition in dozens of different ways
- Solution needed to integrate seamlessly with VA's existing systems
- Maintaining strict data security standards
- Ensuring only current, valid classification codes were used
Sources
- aquia.us — vendor self-report
Similar deployments
Amplitude · Wolfia
4X — throughput of security questionnaires
Vendor-reportedTechnology & software · Document processing · AI agents · Scaled
Adyen · Rossum
70 — automation rate
Vendor-reportedBanking & finance · Document processing · NLP · Scaled · 2023
Skaylink · Moterra
dropped by around 80% — NIS2 preparation time
Vendor-reportedTechnology & software · Document processing · LLM · Scaled
OM Pharma · DrugCard
3 minutes — review time per article
Vendor-reportedPharma & biotech · Document processing · NLP · In production
AMPREP (Amalgamations Repco) · Cimplify.AI
100% — invoice automation
Vendor-reportedManufacturing · Document processing · NLP · In production