ML Engineer – Threat Detection Models

Overview

ML Engineer for Threat Detection Models, building real-time detection for prompts, responses, tool calls, and data security risks.

What you'll do

• Design, train, and evaluate threat detector models using ML and hybrid rule/ML approaches.
• Build and maintain training/evaluation datasets, labeling workflows, and benchmark suites.
• Track detection performance using precision/recall, drift, error analysis, and adversarial robustness.
• Serve models in production with strict latency requirements via Go-based inference services.
• Optimize inference for performance and cost using quantization, distillation, batching, and caching.
• Develop MLOps workflows for reproducible training, registries, monitoring, and safe rollouts.

What you'll need

• 4+ years building and deploying ML models in production, including NLP or LLM-based classification.
• Strong Python skills with PyTorch, Hugging Face Transformers, and scikit-learn.
• Hands-on experience fine-tuning transformer-based models.
• Solid Go skills or backend experience with ability to become productive in Go.
• Experience with low-latency model serving (e.g., ONNX Runtime, TorchServe, Triton, vLLM, or custom).
• Experience with Docker, Kubernetes, and at least one major cloud provider.
• Fluent English with strong written and verbal communication.
• Comfort using AI coding assistants (e.g., Claude Code, Cursor, GitHub Copilot, Codex, or similar).

Nice to have

• Experience building detection systems for prompt injection, jailbreaks, or content safety.
• Experience with DLP/sensitive-data classification (PII, secrets, or source code).
• Familiarity with adversarial ML and model robustness techniques.
• Previous experience in cybersecurity, security tooling, or trust & safety.
• Experience introducing AI-assisted/agentic development workflows across engineering teams.

Details

• Work location: India (Remote-First).
• Employment type: Full-Time; start date: November 2026.