Software Engineer, Sandboxing

Job description

Back to jobs
New
Software Engineer, Sandboxing
San Francisco, CA | New York City, NY
Apply
About Anthropic
Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems.
About the role
Every time Claude runs code, uses a computer, or works as an agent, it does so inside a sandbox. Our Sandboxing team builds the isolation layer behind that work. The same systems run the code Claude writes for users, the environments we train and evaluate models in, and the agents our customers deploy.
These workloads are untrusted by design, and there are a lot of them. Sandboxes need to start in milliseconds, hold up under adversarial code, and scale to a very large number of concurrent sessions. You will work at the boundary between the Linux kernel, virtualization, and the container ecosystem. You will partner with research, product, and security teams to make sure that safe and fast do not trade off against each other.
Key responsibilities
Design, build, and operate the sandboxing infrastructure that runs untrusted, model-generated code and agent workloads at scale
Improve sandbox startup time, density, and throughput, including snapshot and restore, checkpointing, and warm pools
Strengthen isolation across compute, filesystem, and network, including egress controls and per-session credential handling
Investigate and fix performance, reliability, and security issues at the kernel, hypervisor, and container runtime level
Build the monitoring, tracing, and debugging tools that show where time and resources go inside a sandbox
Work with researchers to support reinforcement learning environments and evaluations that need fast, reproducible, isolated execution
Work with product and platform engineers to expose sandboxes as a dependable, well-designed capability for Claude products and customers
Partner with security teams to threat-model the platform and to test, harden, and respond to escape and abuse scenarios
Work with cloud providers on the hardware and platform features that our workloads depend on
Share knowledge on systems programming and Linux internals, and help set technical standards for the team
Minimum qualifications
Software engineering experience in Linux systems programming, kernel development, or another low-level area
Strong programming skills in at least one systems language such as Rust, C, C++, or Go
Working knowledge of virtualization or container isolation technologies such as KVM, QEMU, Firecracker, gVisor, or runc
Understanding of Linux resource management, including namespaces, cgroups, scheduling, and memory management
Experience profiling and debugging performance issues at the system level
Ability to work in unfamiliar codebases and technical areas, and to ship practical solutions with measurable results
Strong communication skills and a commitment to building safe, reliable systems
Preferred qualifications
Experience building or operating a multi-tenant sandbox, serverless, or code execution platform
Experience with microVMs, user-space kernels, or lightweight virtualization, including snapshot and restore
Experience with OS-level security hardening, including seccomp, LSMs, and threat modeling of untrusted code
Experience with eBPF and kernel tracing tools
Experience with Kubernetes and cloud infrastructure at scale (AWS, GCP)
Experience with network isolation, high-performance networking, or building egress proxies and policy layers
Experience with sandboxes for AI agents, computer use, or reinforcement learning environments
Experience with GPU virtualization, accelerators, or performance work on machine learning workloads
Prior experience as a technical lead or mentor
Representative projects
Cutting sandbox cold start from seconds to milliseconds with snapshot-based restore and pre-warmed pools
Designing a network egress policy layer so agents can reach approved services without exposing secrets
Running many thousands of parallel reinforcement learning environments with strong isolation between them
Building tracing tools that attribute latency and resource use inside a running sandbox
Reducing per-sandbox memory overhead to raise density on each host
Hardening the runtime against escape attempts found in internal red-team exercises
The annual compensation range for this role is listed below.
For sales roles, the range provided is the role’s On Target Earnings ("OTE") range, meaning that the range includes both the sales commissions/sales bonuses target and annual base salary for the role.
Annual Salary:
$320,000 - $485,000 USD
Logistics
Minimum education: Bachelor’s degree or an equivalent combination of education, training, and/or experience
Required field of study: A field relevant to the role as demonstrated through coursework, training, or professional experience
Minimum years of experience: Years of experience required will correlate with the internal job level requirements for the position
Location-based hybrid policy: Currently, we expect all staff to be in one of our offices at least 25% of the time. However, some roles may require more time in our offices.
Visa sponsorship: We do sponsor visas! However, we aren't able to successfully sponsor visas for every role and every candidate. But if we make you an offer, we will make every reasonable effort to get you a visa, and we retain an immigration lawyer to help with this.
We encourage you to apply even if you do not believe you meet every single qualification. Not all strong candidates will meet every single qualification as listed. Research shows that people who identify as being from underrepresented groups are more prone to experiencing imposter syndrome and doubting the strength of their candidacy, so we urge you not to exclude yourself prematurely and to submit an application if you're interested in this work. We think AI systems like the ones we're building have enormous social and ethical implications. We think this makes representation even more important, and we strive to include a range of diverse perspectives on our team.
Your safety matters to us. To protect yourself from potential scams, remember that Anthropic recruiters only contact you from @anthropic.com email addresses. In some cases, we may partner with vetted recruiting agencies who will identify themselves as working on behalf of Anthropic. Be cautious of emails from other domains. Legitimate Anthropic recruiters will never ask for money, fees, or banking information before your first day. If you're ever unsure about a communication, don't click any links—visit anthropic.com/careers directly for confirmed position openings.
How we're different
We believe that the highest-impact AI research will be big science. At Anthropic we work as a single cohesive team on just a few large-scale research efforts. And we value impact — advancing our long-term goals of steerable, trustworthy AI — rather than work on smaller and more specific puzzles. We view AI research as an empirical science, which has as much in common with physics and biology as with traditional efforts in computer science. We're an extremely collaborative group, and we host frequent research discussions to ensure that we are pursuing the highest-impact work at any given time. As such, we greatly value communication skills.
The easiest way to understand our research directions is to read our recent research. This research continues many of the directions our team worked on prior to Anthropic, including: GPT-3, Circuit-Based Interpretability, Multimodal Neurons, Scaling Laws, AI & Compute, Concrete Problems in AI Safety, and Learning from Human Preferences.
Come work with us!
Anthropic is a public benefit corporation headquartered in San Francisco. We offer competitive compensation and benefits, optional equity donation matching, generous vacation and parental leave, flexible working hours, and a lovely office space in which to collaborate with colleagues. Guidance on Candidates' AI Usage: Learn about our policy for using AI in our application process.
Create a Job Alert
Interested in building your career at Anthropic? Get future opportunities sent straight to your email.
Create alert
Apply for this job
*
indicates a required field
First Name*
Last Name*
Email*
Phone
Country
Phone
Resume/CV
AttachAttach
Dropbox
Google Drive
Enter manuallyEnter manually
Accepted file types: pdf, doc, docx, txt, rtf
(Optional) Personal Preferences
How do you pronounce your name?
Website
Are you open to working in-person in one of our offices 25% of the time?*
Select...
When is the earliest you would want to start working with us?
Do you have any deadlines or timeline considerations we should be aware of?
AI Policy for Application*
Select...
We believe that AI will have a transformative impact on the world, and we’re seeking exceptional candidates who collaborate thoughtfully with Claude to realize this vision. At the same time, we want to understand your unique skills, expertise, and perspective through our hiring process. We invite you to review our AI partnership guidelines for candidates and confirm your understanding by selecting “Yes.”
Why Anthropic?*
Why do you want to work at Anthropic? (We value this response highly - great answers are often 200-400 words.)
Do you require visa sponsorship? *
Select...
Will you now or will you in the future require employment visa sponsorship to work in the country in which the job you're applying for is located? *
Select...
Additional Information
Add a cover letter or anything else you want to share.
LinkedIn Profile
Please ensure to provide either your LinkedIn profile or Resume, we require at least one of the two.
Are you open to relocation for this role? *
Select...
What is the address from which you plan on working? If you would need to relocate, please type "relocating".
Have you ever interviewed at Anthropic before?*
Select...
Please read the arbitration agreement below*
Select...
Agreement to Arbitrate
I understand and agree that both Anthropic and I waive our respective rights to trial by jury in connection with any claims covered by this agreement to arbitrate. Instead, any disputes arising out of or related to my application for employment with Anthropic shall be settled by final and binding arbitration pursuant to the Federal Arbitration Act. I understand and agree that Anthropic’s employees/agents, Anthropic’s clients, and their employees/agents are third-party beneficiaries to this agreement.
The arbitration shall be conducted in accordance with the Employment Arbitration Rules & Procedures of Judicial Arbitration and Mediation Services, Inc. (“JAMS Rules”), available via the internet at https://www.jamsadr.com/rules-employment-arbitration or by searching for “JAMS Employment Arbitration Rules” in an online search service such as www.google.com. To the extent the JAMS Rules are inconsistent with the terms of this Agreement, this Agreement shall govern to the extent permitted by law. The parties will select one neutral arbitrator, who will have the authority to award all relief in law or equity that is requested by the parties and supported by credible, relevant, and admissible evidence. The arbitrator shall allow adequate discovery and issue a written, signed, and reasoned award. Judgement may be entered on the arbitrator’s award in any court having jurisdiction, and the award shall be subject to correction, confirmation, or vacation as provided by the applicable law concerning judicial review of arbitration awards. Anthropic will bear the costs that are particular to the arbitration, such as the arbitrator fees. The arbitrator will apply the substantive law of the state in which I sought employment with Anthropic. The arbitrator will determine the arbitration venue, based on fairness and convenience to the parties and witnesses, unless otherwise mutually agreed upon by the parties. The Party initiating a claim under this Agreement must make a written demand for arbitration to the other Party. The demand for arbitration shall identify the claims asserted, the facts upon which such claims are based, and any relief or remedy sought. Written demand for arbitration to Anthropic must be sent to Anthropic PBC at arbdemands@anthropic.com. Anthropic’s demand for arbitration shall be sent to the most recent mailing address and/or email address that Anthropic has on file for you.
This agreement to arbitrate does not apply to claims for workers’ compensation, unemployment compensation benefits, claims or charges before any administrative agency having jurisdiction over such claims, or any other claim that is not subject to arbitration under federal law. If any claims are deemed non-arbitrable, then those claims will be stayed until resolution of any arbitrable claims.
I understand and agree that, except where prohibited by federal law, all claims subject to this agreement to arbitrate must be pursued on an individual basis, and that both I and Anthropic waive any right to bring or be a party to any class or collective action. I agree that any issues pertaining to the enforceability, application or validity of this agreement to arbitrate shall be decided only by a court of competent jurisdiction and not by an arbitrator. This agreement to arbitrate applies only in the United States.
Agreement to Arbitrate*
Select...
Voluntary Self-Identification
For government reporting purposes, we ask candidates to respond to the below self-identification survey.
Completion of the form is entirely voluntary. Whatever your decision, it will not be considered in the hiring
process or thereafter. Any information that you do provide will be recorded and maintained in a
confidential file.
As set forth in Anthropic’s Equal Employment Opportunity policy,
we do not discriminate on the basis of any protected group status under any applicable law.
Gender
Select...
Are you Hispanic/Latino?
Select...
Race & Ethnicity Definitions
If you believe you belong to any of the categories of protected veterans listed below, please indicate by making the appropriate selection.
As a government contractor subject to the Vietnam Era Veterans Readjustment Assistance Act (VEVRAA), we request this information in order to measure
the effectiveness of the outreach and positive recruitment efforts we undertake pursuant to VEVRAA. Classification of protected categories
is as follows:
A "disabled veteran" is one of the following: a veteran of the U.S. military, ground, naval or air service who is entitled to compensation (or who but for the receipt of military retired pay would be entitled to compensation) under laws administered by the Secretary of Veterans Affairs; or a person who was discharged or released from active duty because of a service-connected disability.
A "recently separated veteran" means any veteran during the three-year period beginning on the date of such veteran's discharge or release from active duty in the U.S. military, ground, naval, or air service.
An "active duty wartime or campaign badge veteran" means a veteran who served on active duty in the U.S. military, ground, naval or air service during a war, or in a campaign or expedition for which a campaign badge has been authorized under the laws administered by the Department of Defense.
An "Armed forces service medal veteran" means a veteran who, while serving on active duty in the U.S. military, ground, naval or air service, participated in a United States military operation for which an Armed Forces service medal was awarded pursuant to Executive Order 12985.
Veteran Status
Select...
Submit application