AI Influence Operation Detection Developer

Corvex Intelligence

Remote · $160k – $200k / 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

Coordinated inauthentic behavior — networks of fake accounts amplifying narratives, astroturfing campaigns, state-sponsored disinformation — is measurable if you know what to look for. Posting pattern analysis, network graph anomalies, linguistic fingerprinting, and narrative coordination signals each tell part of the story. Combined, they make influence operations detectable before they achieve their objectives. We build detection infrastructure for government clients, election integrity organizations, and large social platforms. The ML challenge is adversarial: operators learn from public detection research and adapt. Models that worked 18 months ago are partially evaded today. Staying ahead requires continuously evolving the detection approach. You'll own the ML detection pipeline: developing new features from behavioral and linguistic signals, training classifiers on labeled datasets from documented influence operations, and building the adversarial robustness testing that evaluates whether new evasion techniques defeat current detectors before they are deployed in the wild.

Fully remote.

Responsibilities

Develop behavioral and linguistic features for coordinated inauthentic behavior detection
Train and evaluate influence operation classifiers on labeled historical datasets
Build adversarial robustness testing using known evasion techniques
Design detection systems that degrade gracefully as adversaries adapt
Collaborate with government and platform clients on operational deployment requirements
Contribute threat intelligence research on emerging influence operation techniques

Requirements

5+ years in ML engineering with adversarial ML, disinformation, or social network analysis
Experience with graph-based features and network anomaly detection
Strong Python and NLP skills for linguistic fingerprinting
Understanding of adversarial robustness and concept drift in adversarial settings
Security clearance eligibility preferred