Real-Time Fraud & Abuse Detection ML Engineer

Telnyx LLC; Tandil; On-site; ARS 1,800,000 - 2,400,000; Full time

Job summary

Telnyx is seeking a Machine Learning Engineer on the Messaging team to own real-time decisioning for outbound messages, building fraud and abuse detection across phishing, brand impersonation, and scams.

You will join at an inflection point as we rebuild our detection stack on a modern decision-engine architecture, enabling inline hot-path decisioning and a full audit ledger. Grow with a financially stable company focused on learning and growth.

Qualifications

• 3+ years of software or ML experience.
• Proficiency in async Python web frameworks (FastAPI).
• Production experience with text classification systems for fraud/abuse.
• Rigorous evaluation methodology: labeled corpora, holdout sets, precision/recall, calibration.
• Experience using LLMs as classifiers within a larger decision pipeline.
• Experience deploying and maintaining ML models in production.
• Excellent communication and cross-functional collaboration.

Responsibilities

• Design, train, evaluate, and operate text classifiers end-to-end, including data labeling and drift monitoring.
• Run shadow-mode evaluations against fraud corpora and legitimate controls; report precision, recall, calibration.
• Build detection against adversarial content and evasion attempts in text and URLs.
• Serve detection models inline with low latency and controlled fail-open/close behavior.
• Support human-in-the-loop fraud review; contribute to tooling and signals for training data.
• Collaborate with fraud operations, trust & safety, and cross-functional engineering teams.
• Participate in code reviews and raise code quality and standards.

Skills

Years of experience

Async Python (FastAPI)

Text classification in fraud/abuse

Evaluation methodology

LLMs in decision pipelines

Production ML deployment

Communication & collaboration

Tools

ClickHouse

PostgreSQL

Redis

Kafka

Docker

Kubernetes

Prometheus

Ray

Job description

Telnyx is seeking a Machine Learning Engineer on the Messaging team to own real-time decisioning for outbound messages, building fraud and abuse detection across phishing, brand impersonation, and scams.

You will join at an inflection point as we rebuild our detection stack on a modern decision-engine architecture, enabling inline hot-path decisioning and a full audit ledger. Grow with a financially stable company focused on learning and growth.