Skip to content

Commit

Permalink
Merge pull request #1 from terraform/dubos306-patch-2
Browse files Browse the repository at this point in the history
Review QUANTM
  • Loading branch information
badra001 committed May 5, 2026
2 parents ec20db3 + 87fa996 commit 41dfb3b
Showing 1 changed file with 59 additions and 0 deletions.
59 changes: 59 additions & 0 deletions inventory/20260504-1.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,59 @@
template_date: 2026-05-05
template_version: 1.0.2
basic:
date: 2026-05-04
version: 1.0.0
contact_info:
- name: Sreenivas Karpurapu
division: DITD
mail: sreenivas.karpurapu@census.gov
- name: Ama Danso
division: DITD
mail: Ama.A.Danso@census.gov

bedrock_information:
company_name: U.S. Census Bureau
website_url: https://census.gov/
industry: Government
intended_users:
internal_employees: true
external_user: false
use_case_description: |
QUANTM's AI Enhanced Clerical Matching Solution evaluates the application of large language models to support
In Field Enumeration (IFE) and Census Data Quality Assurance (CDQA) by performing intelligent matching of complex records,
reducing manual clerical effort while producing transparent, explainable decisions consistent with expert human judgment.
bedrock_models:
- anthropic.claude-4-5-sonnet

federal_standards_information:
id: 20260504-1
project_name: QUANTM
project_summary: |
As part of the quality operations of the US Census Bureau, computer matching and clerical operations are employed. These efforts help in the assessment
of the quality of data collection. The existing operations have constraints in their ability to accurately match and compare complex datasets, resulting in high number
of cases being flagged or removed. This process is time-intensive and expensive, creating a significant burden on human resources. With the advanced AI capabilities
available in the industry, QUANTM is assessing their feasibility to support the quality assessment efforts and reduce the clerical footprint. The bureau aims to explore
whether large language model (LLM) technology can replicate and improve upon the reasoning processes used by human clerks.
development_stage: active
data_sensitivity: |
Title 13. Demographic data from census survey responses
dms_project_number: 7535836
cms_project_number: null

deployment_details:
crf_number: NUMBER
finops_project_number: 1
finops_project_name: edl_ditd
accounts:
- account_id: 001502248272
account_alias: edl-addcp-prod-gov
regions:
- us-gov-west-1
environment: prod
commercial_account_id: 563127383709
commercial_account_alias: edl-addcp-prod-ew

history:
- date: 20260505
comment: created

0 comments on commit 41dfb3b

Please sign in to comment.