Latest AI JobsBrowse Jobs
Anthropic logo

ML/Research Engineer, Safeguards

AnthropicSan Francisco, CA | New York City, NYAI Research & Engineering
Apply on company site
Salary
Not listed
Job type
Full-time
Posted
9 Oct 2025
Status
Open · checked 7d ago

About this role

Anthropic is hiring an ML/Research Engineer, Safeguards to join the AI Research and Engineering team in San Francisco, CA or New York City, NY. This position is well-suited for professionals eager to protect artificial intelligence products from misuse and ensure model safety as capabilities advance.

Role Overview

The Safeguards ML team focuses on detecting and mitigating the misuse of artificial intelligence models, ranging from individual policy breaches to coordinated attacks. Professionals in this position build systems that identify harmful behaviors, safeguard user well-being, and maintain appropriate model responses across diverse contexts. This vital work directly supports Anthropic's Responsible Scaling Policy commitments.

Key Responsibilities

Design and deploy classifiers to identify misuse and anomalous behaviors at scale, utilizing synthetic data pipelines and representative evaluations.
Construct monitoring systems to catch multi-exchange harms like coordinated cyber attacks and influence operations by aggregating signals across contexts.
Assess and enhance the safety of agentic products through specialized threat models, testing environments, and targeted mitigations for prompt injection attacks.
Perform exploratory investigations regarding automated red-teaming, adversarial robustness, and related methods to uncover and prevent misuse.

Requirements

Possess at least 4 years of professional or academic background in applied research, machine learning engineering, or research engineering.
Demonstrate proficiency in Python along with proven capability in building machine learning systems.
Show comfort handling the entire research-to-deployment pipeline, transitioning smoothly from experimental phases to production environments.
Bring strong communication skills capable of explaining intricate technical ideas to non-technical individuals.
Hold a minimum of a Bachelor's degree or equivalent combination of education, training, and experience in a relevant field of study.

Nice to Have

Familiarity with language modeling and transformers.
Background in building classifiers, anomaly detection architectures, or behavioral machine learning.
Experience dealing with adversarial machine learning or red-teaming.
Knowledge of interpretability methods or probes.
Exposure to reinforcement learning or high-performance, large-scale machine learning systems.

Skills and Tools

Python
Machine Learning Systems
Synthetic Data Pipelines
Classification and Anomaly Detection
Adversarial Robustness and Red-Teaming

Salary and Benefits

Annual salary range of $350,000 to $500,000 USD.
Competitive compensation and benefits package.
Optional equity donation matching.
Generous vacation time and parental leave.
Flexible working hours.
Visa sponsorship provided with immigration lawyer support when needed.

About the Company

Anthropic operates as a public benefit corporation focused on creating reliable, interpretable, and steerable artificial intelligence systems that benefit society. The organization approaches AI research as an empirical science, valuing large-scale collaborative projects over isolated puzzles. Staff members operate under a location-based hybrid policy requiring presence in a company office at least 25% of the time.

How to Apply

Interested candidates should visit the official Anthropic careers page to submit an application for this position.

Skills and tech stack

Python

Questions candidates ask

What is the location for the ML/Research Engineer, Safeguards role?

The position is based in San Francisco, CA or New York City, NY, with a hybrid office policy requiring staff to be in the office at least 25% of the time.

What is the salary range for this position at Anthropic?

The annual salary range for this role is between $350,000 and $500,000 USD.

How many years of experience are required to apply?

Candidates must have 4 or more years of experience in machine learning engineering, research engineering, or applied research.

Does Anthropic offer visa sponsorship for this role?

Yes, Anthropic sponsors visas and retains an immigration lawyer to assist candidates receiving an offer, though successful sponsorship cannot be guaranteed for every single case.

What programming language is required?

Proficiency in Python is required for this role.

Ready to apply?

The application is on the official Anthropic careers page.

Apply on company site