Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 

Repository files navigation

Parche Agent

Agentic data intelligence — ingest structured or unstructured records, synthesize insights, and answer questions with citations, evaluation, and cost observability.

Problem

Teams pay for data access but still spend hours manually reading records to find patterns. This turns raw records into synthesized, queryable intelligence.

First use case

Import/competitor intelligence: ingest import records → summarize who is importing what, surface trends, answer natural-language questions.

Roadmap

  • Phase 1: Core agent loop (tool selection + memory)
  • Phase 2: Retrieval (chunking, embeddings, vector store)
  • Phase 3: FastAPI service (ingest + query endpoints)
  • Phase 4: Evaluation + observability (LLM-as-judge, cost logging)
  • Phase 5: Docker + cloud deploy (live URL)
  • Phase 6: Write-up (design decisions + tradeoffs)

Tech

Python · FastAPI · [vector store TBD] · Docker · [cloud TBD]

About

Agentic data intelligence — project to create an agent campable of ingesting structured or unstructured records, synthesize insights, and answer questions with citations, evaluation, and cost observability.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages