Finance ops · Production

FloQast builds an AI-powered accounting transformation solution with Anthropic's Claude 3 on Amazon Bedrock

The problem

As organizations scale, accounting operations become exponentially complex, requiring meticulous documentation, categorization, and reconciliation of large transaction volumes. The final 20% of financial processes—intricate and organization-specific—still required significant manual intervention, consuming accounting teams' hours on repetitive spreadsheet work.

Workflow diagram · grounded in source
1
Transaction data ingestion
trigger
“Transaction data is gathered from bank statements and enterprise resource planning (ERP) systems.”
2
Accountant selects and requests rule
trigger
“An accountant will select specific transactions in both systems and choose Generate AI Rule.”
3
AI generates matching rule text
ai_action
“Based on the selected transactions, text is generated”
4
Accountant reviews generated rule
human_review
“the accountant has the option to either accept the generated text or edit the text”
5
Rule saved and applied
output
“The accountant chooses Save and apply to generate a rule in coded format that is further used to find additional matches”
6
Audit documents uploaded to S3
trigger
“Users upload supporting documents that provide audit evidence into a secure Amazon Simple Storage Service (Amazon S3) bucket.”
7
Textract extracts document data
ai_action
“extracts the data from the documents”
8
LLM generates audit annotations
ai_action
“The LLM runs the audit rules against the extracted data and generates an annotation for each audit rule, including pass/fail details of the audit rule.”
9
Guardrails filter annotation results
validation
“Annotation results are filtered using Amazon Bedrock Guardrails to enhance content safety and privacy in generative AI applications.”
Reported outcome

FloQast's AI-powered accounting transformation solution delivered a 38% reduction in reconciliation time, a 23% decrease in audit process duration and discrepancies, and a 44% improvement in workload management, saving accounting teams countless hours and enabling focus on higher-value activities.

Reported metrics
Reconciliation time reduction38%
Audit process duration and discrepancies decrease23%
Workload management improvement44%
Accounting team time savingssaved accounting teams countless hours
Reported stack
Amazon BedrockClaude 3.5 SonnetAmazon TextractAmazon S3AWS Step FunctionsMongoDBAmazon Bedrock GuardrailsAmazon Bedrock AgentsRAG
Source
https://aws.amazon.com/blogs/machine-learning/floqast-builds-an-ai-powered-accounting-transformation-solution-with-anthropics-claude-3-on-amazon-bedrock?tag=soumet-20
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

FloQast's AI-powered accounting transformation solution delivered a 38% reduction in reconciliation time, a 23% decrease in audit process duration and discrepancies, and a 44% improvement in workload management, savin…

What tools did this team use?

Amazon Bedrock, Claude 3.5 Sonnet, Amazon Textract, Amazon S3, AWS Step Functions, MongoDB, Amazon Bedrock Guardrails, Amazon Bedrock Agents, RAG.

What results were reported?

Reconciliation time reduction: 38%; Audit process duration and discrepancies decrease: 23%; Workload management improvement: 44%; Accounting team time savings: saved accounting teams countless hours (source-reported, not independently verified).

How is this finance ops AI workflow structured?

Transaction data ingestion → Accountant selects and requests rule → AI generates matching rule text → Accountant reviews generated rule → Rule saved and applied → Audit documents uploaded to S3 → Textract extracts document data → LLM generates audit annotations → Guardrails filter annotation results.

Related finance ops cases