Founding AI Scientist - Security for AI

Deca Talent; San Francisco (CA); On-site; USD 180,000 - 280,000; Full time

Job summary

Deca Talent is partnering with a well-funded, early-stage AI security startup to hire a Founding AI Scientist / ML Research Engineer. You will bridge research and engineering, turning modern ML/LLM research into production-ready systems that operate with enterprise data.

You’ll develop LLM and ML systems for security apps, design evaluation frameworks, and explore new reasoning and decision-making approaches with LLMs while shaping the platform’s technical direction in its seed phase.

Qualifications

• Strong experience with LLMs and modern machine learning.
• Hands-on experience with model evaluations / LLM evals.
• Strong ML engineering capabilities alongside research experience.
• Comfortable building production-quality systems, not just prototypes.
• Able to work independently and take ownership of significant technical problems.
• Experience moving research into real-world products.

Responsibilities

• Developing LLM and ML systems for AI security applications
• Designing and building rigorous model evaluation / eval frameworks
• Researching new approaches to reasoning, classification and decision-making with LLMs
• Working with real-world enterprise data, identity, access and behavioral signals
• Helping shape the technical direction of the AI platform at an early-stage company

Skills

LLMs

ML Engineering

Model Evaluations

Production Systems

Job description

Founding AI Scientist / ML Research Engineer

I’m partnering with a well-funded, early-stage AI security startup that is building a new category of infrastructure for controlling how AI assistants and autonomous agents interact with enterprise data. The company has recently shipped its MVP and is now working with early design partners as it moves into its next stage of commercialization.

They’re looking for an AI Scientist / ML Research Engineer who can operate across research and engineering - someone who can take modern ML/LLM research, turn it into working systems, and ultimately ship it into production.

What you’ll work on:

• Developing LLM and ML systems for AI security applications
• Designing and building rigorous model evaluation / eval frameworks
• Researching new approaches to reasoning, classification and decision-making with LLMs
• Working with real-world enterprise data, identity, access and behavioral signals
• Helping shape the technical direction of the AI platform at an early-stage company

What they’re looking for:

• Strong experience with LLMs and modern machine learning
• Hands‑on experience with model evaluations / LLM evals
• Strong ML engineering capabilities alongside research experience
• Comfortable building production‑quality systems, not just prototypes
• Able to work independently and take ownership of significant technical problems
• Experience moving research into real-world products

Nice to have:

• AI safety experience
• Experience at an AI research or safety lab
• Background in AI security, trust & safety, or related areas