← Cases
United StatesProfessional services2025 – ongoing

Aura Intelligence

Job title and industry classification with Claude in Amazon Bedrock

Aura Intelligence cut job title classification from 2-3 months to 30 minutes with Claude

WorkingVendor case studyVendor ↗

Problem

Aura Intelligence sells workforce data and insights to consulting firms, private equity firms and hedge funds. Classifying job titles and roles across industries and languages relied on manual lookups and fuzzy matching.

Approach

Aura used Claude in Amazon Bedrock to classify titles and roles across industries and languages. It also uses Claude for report generation, anomaly detection in workforce trends, sentiment analysis of company reviews and translation.

Results

2-3 months to 30 minutestitle classification timeVendor case study[1]
94%accuracy in overall classification tasksVendor case study[1]
under 8%rate of unclassified data across industriesVendor case study[1]
over 200 milliontitles and industry pairings processed across multiple languagesVendor case study[1]

Takeaway

Classification that took months now runs in minutes. Aura did not say how it tested the accuracy.

Categories

Business function
Production and quality
AI technology
Generative AI
Evidence
Vendor case study

Sources

Source summaries

[1]Anthropic28 Sep 2025

Anthropic describes how Aura Intelligence uses Claude in Amazon Bedrock to classify job titles and roles and to support reporting and analysis. It reports 94% overall accuracy, classification time falling from 2-3 months to 30 minutes, under 8% unclassified data, and over 200 million title and industry pairings processed. The page carries no date; the Internet Archive first saved it on September 28, 2025.

Full details are in the original sources.

Added 4 Nov 2026 · Updated 4 Nov 2026
Open case page ↗