Machine Learning Engineer: LLM Interpretability & Systems ($175K – $250K Equity) at CTGT, Inc.

This is a job that Jill, our AI Recruiter, is recruiting for on behalf of one of our customers. She will pick the best candidates from Jack’s network.

Job Title

Machine Learning Engineer: LLM Interpretability & Systems

Salary

$175K — $250K Equity

Company Description

CTGT, Inc. is a venture-backed startup born out of Stanford University research, providing a deterministic governance layer for enterprise AI. Backed by Google’s Gradient Ventures, General Catalyst, and Y Combinator, CTGT, Inc. builds the control plane that enables Fortune 500 institutions to deploy generative AI workflows with auditability and confidence.

Job Description

As a Machine Learning Engineer at CTGT, Inc., you will work deep within the model stack to make generative AI deterministic and reliable. You’ll apply mechanistic interpretability techniques to model weights and activations, building the policy engine that enforces real-time governance for high-stakes enterprise applications at the edge of AI research.

Location

San Francisco, USA

Why this role is remarkable

• Bridge the gap between frontier AI research and real-world production by implementing mechanistic interpretability techniques like activation patching and control vectors.
• Join a high-growth team backed by elite investors including Google’s Gradient Ventures, General Catalyst, and Y Combinator with significant equity upside.
• Develop the core deterministic governance layer required by the world’s largest institutions to safely deploy generative AI at scale.

What You Will Do

• Operationalize mechanistic interpretability research into production-ready code that improves model behavior through direct internal interventions.
• Design and optimize feature-level intervention systems that enable real-time, auditable policy enforcement during model inference.
• Build robust evaluation and deployment loops to ensure model changes are reliably shipped into complex enterprise VPC environments.

The ideal candidate

• Possesses deep expertise in Transformer architectures, PyTorch internals, and the mathematical foundations required for advanced deep learning optimization.
• Demonstrates a proven track record of training or fine-tuning models beyond simple augmentation, specifically probing the mechanics of model cognition.
• Exhibits strong technical ownership with the ability to translate academic papers into high-performance Python, Rust, or TypeScript implementations.

We never post fake jobs

This isn’t a trick. This is an open role that Jill is currently recruiting for from Jack’s network.