AI Reliability & Red Teaming Engineer

Role Overview

​Your mission is to break our models before someone else does. You will lead the technical stress-testing of our AI systems, focusing on Adversarial Robustness, Bias detection, and Hallucination mitigation.

• ​Key Responsibilities
• ​Red Teaming: Design and execute adversarial attacks (prompt injection, data poisoning) against LLMs and predictive models.
• ​Automated Monitoring: Build and maintain the "Continuous Control" stack using tools like Evidently AI or WhyLabs.
• ​Bias Remediation: Partner with Data Scientists to implement algorithmic fairness techniques when bias is detected.
• ​Qualifications
• ​Technical Stack: Expert Python skills; deep experience with PyTorch, TensorFlow, and MLOps pipelines.
• ​Education: BS/MS in Data Science, Mathematics, or Cybersecurity.
• ​Experience: 4+ years in ML Engineering with a focus on model security or reliability.
• Originally posted on Himalayas