Natnael Alemseged
AboutProjectsTestimonialsWork Experience
© 2026 Natnael Alemseged. All Rights Reserved.
Secure Agent Protocol // Latency Critical // Addis Ababa

Axiom Ledger

Event-Sourced AI Workflow Engineer

Event-sourced lending pipeline for document intake, extraction, credit analysis, fraud, compliance, and decision orchestration over an append-only ledger.

"Modeled lending decisions as replayable event streams so every agent action has an audit trail."
Axiom Ledger event-sourced lending pipeline
Click to Zoom
Append-only loan event streams powering document, credit, fraud, compliance, and orchestration agents

Problem

AI-assisted lending workflows need strict traceability across document processing, credit reasoning, fraud checks, compliance blocks, and final decision outcomes.

Solution

Created a canonical event schema, synthetic company/document generator, in-memory and PostgreSQL event stores, multi-agent pipeline phases, optimistic concurrency tests, projections, and refinery-backed document extraction.

Deep Dive

What It Tracks

Axiom Ledger represents lending workflows as streams of immutable events. Document packages, extraction results, credit analysis, fraud checks, compliance decisions, and orchestrator outcomes can all be replayed from the ledger.

Engineering Highlights

  • •45 event types: typed envelopes and schemas define the domain contract.
  • •Five-agent pipeline: document, credit, fraud, compliance, and decision orchestration phases.
  • •Document Refinery integration: financial facts can be extracted from PDFs and normalized into ledger events.

Tech Stack

PythonPostgreSQLEvent SourcingPydanticpytestDocker

Tags

#Event Sourcing#Fintech#Audit Trail#Multi-Agent Systems
View GitHub Repo

More Software Software

Case studies in similar engineering domains.

Brownfield Cartographer

→

Multi-agent codebase cartography tool that analyzes local or GitHub repositories with Surveyor and Hydrologist agents to produce module graphs and data lineage artifacts.

Data Contract Enforcer

→

Schema integrity and lineage attribution system that turns inter-system dependencies into formal contracts, detects schema/type/statistical drift, and reports downstream blast radius.

Document Intelligence Refinery

→

PDF triage and extraction pipeline that detects document origin, layout, and domain, escalates extraction strategies by confidence, builds PageIndex trees, and answers with provenance chains.