AI Knowledge Graph Conflict Resolution Developer

Altrix Systems

Remote · $100k – $130k / year · Full-time · Posted 4 months ago

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

About the role

A knowledge graph built from multiple sources will contain conflicts. Two news articles say the same person was born in different cities. A corporate registry says a company was founded in 2008; a Wikipedia scrape says 2009. A scientific database records a drug as approved for condition A; a regulatory filing says the approval is pending. These conflicts are a signal — they reveal data quality problems, source disagreements, or genuinely uncertain facts — but only if you can detect and surface them systematically. Our knowledge graph platform serves 28 enterprise clients in finance, healthcare, and legal sectors whose business decisions depend on the accuracy of the knowledge we maintain. Conflict detection is a critical reliability feature: before any fact in our graph is surfaced to a downstream application, we need to know whether it's unambiguously supported, contested across sources, or temporally superseded. You'll build the conflict resolution system: detecting contradictory assertions across source documents, estimating source credibility weights, implementing resolution strategies for different conflict types, and building the confidence scoring that expresses how certain we are about each contested fact. The system serves 180 million graph assertions and needs to scale.

Fully remote.

Responsibilities

Build conflict detection pipeline identifying contradictory assertions across 180M knowledge graph entries
Implement source credibility weighting for resolution prioritization across heterogeneous data sources
Design resolution strategies for temporal, categorical, and numerical conflict types
Create confidence scoring for contested facts based on source agreement and credibility
Build conflict audit trail documenting resolution decisions for client inspection
Develop conflict resolution accuracy evaluation using domain expert annotation as ground truth

Requirements

3+ years in knowledge graph engineering, information extraction, or data quality systems
Strong Python, SQL, and NLP skills with graph data familiarity
Experience with entity resolution, data fusion, or multi-source information reconciliation
Understanding of knowledge representation and assertion confidence modeling
Ability to design systems that scale to hundreds of millions of assertions