
Machine Learning Engineer ($175K – $250K + Equity) at Stanford-born AI governance startup
Jack & Jill · San Francisco, CA
- On site
- Full-time
- $250,000 / year
- San Francisco, CA
Job highlights
- Build AI governance layer for enterprise.
- Use mechanistic interpretability on models.
- Work with Stanford-born, elite-backed startup.
- Develop core 'Policy Engine' for AI deployment.
- Translate research into production code.
About the role
Machine Learning Engineer: LLM Interpretability & Systems
Company: Jack & Jill (recruiting for a Stanford-born AI governance startup)
Salary: $175K – $250K + 0.5% – 1% Equity
Location: San Francisco, USA
About the Role
You will operate deep within the model stack to build the deterministic governance layer for enterprise AI. By leveraging mechanistic interpretability, you'll work directly with model internals—weights and activations—to enforce policy and prevent drift. This role transforms frontier research into production systems that make LLMs reliable for Fortune 500 institutions.
Why this role is remarkable
- Work at the intersection of frontier AI research and production environments, influencing the mechanics of model cognition.
- Join a high-pedigree team born out of Stanford research, backed by elite investors including Google’s Gradient Ventures and Y Combinator.
- Drive massive impact by building the core "Policy Engine" that enables the world's most important institutions to deploy generative AI with confidence.
What You Will Do
- Implement techniques like activation patching and control vectors for targeted, repeatable improvements in model output.
- Design and optimize feature-level intervention systems for deterministic policy enforcement at inference time.
- Build evaluation and deployment loops for shipping interpretability-based changes reliably into complex enterprise environments.
The Ideal Candidate
- Possesses a deep mathematical foundation in Transformer architectures and PyTorch internals, with experience training or fine-tuning models.
- Demonstrates the ability to translate academic papers on mechanistic interpretability into robust, production-ready code.
- Exhibits an ownership mindset and technical curiosity, driven to solve the challenge of making non-deterministic models auditable and controllable.
Key skills/competency
- Machine Learning Engineer
- LLM Interpretability
- Transformer Architectures
- PyTorch
- Mechanistic Interpretability
- Production Systems
- AI Governance
- Activation Patching
- Control Vectors
- Generative AI
Skills & topics
- Machine Learning Engineer
- LLM Interpretability
- AI Governance
- PyTorch
- Transformer Architectures
- Mechanistic Interpretability
- Production Systems
- Generative AI
- Stanford
- Startup
How to get hired
- Tailor your resume: Highlight experience with Transformer architectures, PyTorch, and mechanistic interpretability.
- Showcase research translation: Provide examples of turning academic papers into production code.
- Emphasize ownership: Demonstrate technical curiosity and a drive for auditable models.
- Prepare for technical deep dives: Be ready to discuss model internals, weights, and activations.
Technical preparation
Master PyTorch and Transformer internals.,Study mechanistic interpretability research papers.,Practice implementing activation patching.,Build and deploy inference-time systems.
Behavioral questions
Describe a complex research paper you implemented.,How do you ensure model auditability?,Share an ownership mindset example.,How do you prevent AI policy drift?
Frequently asked questions
- What is the primary focus of the Machine Learning Engineer role at this Stanford-born AI startup?
- The Machine Learning Engineer role focuses on building a deterministic governance layer for enterprise AI, specifically leveraging mechanistic interpretability to enforce policy and prevent drift in LLMs.
- What kind of AI research and production experience is required for the Machine Learning Engineer position?
- The role requires experience at the intersection of frontier AI research and production environments, moving beyond simple prompting to influence model cognition and translating academic interpretability papers into production code.
- What specific techniques will a Machine Learning Engineer use in this role?
- A Machine Learning Engineer will implement techniques such as activation patching and control vectors, and design feature-level intervention systems for deterministic policy enforcement at inference time.
- What is the significance of the 'Policy Engine' mentioned in the job description?
- The 'Policy Engine' is the core system this role will build, enabling major institutions to deploy generative AI with confidence by ensuring its reliability and controllability.
- What level of mathematical and programming foundation is expected for this Machine Learning Engineer role?
- Candidates should have a deep mathematical foundation in Transformer architectures and PyTorch internals, with practical experience in training or fine-tuning models.
- How does the Jack & Jill recruitment process work for this Machine Learning Engineer job?
- Jill, an AI Recruiter for the company, recruits from Jack's network. If Jack (your AI career coach) identifies you as a strong candidate, an introduction will be made to the company.
- What makes this Machine Learning Engineer role unique compared to others in AI governance?
- This role is unique because it focuses on influencing the mechanics of model cognition through interpretability, building a core governance layer for enterprise AI, and working with a high-pedigree, well-funded startup.
- What are the investment backing and team pedigree for this AI startup?
- The startup is Stanford-born and backed by elite investors including Google’s Gradient Ventures, General Catalyst, and Y Combinator, with a team originating from Stanford research.
Similar roles
Open positions we recommend based on this role.