AI Safety Researcher

About TaskUs: TaskUs is a provider of outsourced digital services and next-generation customer experience to fast-growing technology companies, helping its clients represent, protect and grow their brands. Leveraging a cloud-based infrastructure, TaskUs serves clients in the fastest-growing sectors, including social media, e-commerce, gaming, streaming media, food delivery, ride-sharing, HiTech, FinTech, and HealthTech. 

The People First culture at TaskUs has enabled the company to expand its workforce to approximately 45,000 employees globally. Presently, we have a presence in twenty-three locations across twelve countries, which include the Philippines, India, and the United States.

It started with one ridiculously good idea to create a different breed of Business Processing Outsourcing (BPO)! We at TaskUs understand that achieving growth for our partners requires a culture of constant motion, exploring new technologies, being ready to handle any challenge at a moment’s notice, and mastering consistency in an ever-changing world.

What We Offer: At TaskUs, we prioritize our employees' well-being by offering competitive industry salaries and comprehensive benefits packages. Our commitment to a People First culture is reflected in the various departments we have established, including Total Rewards, Wellness, HR, and Diversity. We take pride in our inclusive environment and positive impact on the community. Moreover, we actively encourage internal mobility and professional growth at all stages of an employee's career within TaskUs. Join our team today and experience firsthand our dedication to supporting People First.

Frontier AI systems evolve daily, and real‑world safety has to keep pace. TaskUs powers applied AI safety for top model builders and application developers, and we’re scaling our research bench to stay ahead of the curve.
As an AI Safety Researcher you’ll design experiments, prototype interventions, and publish findings that turn cutting‑edge theory into practical tools for our clients.
 

The impact you’ll make:

• Advance the science: Contribute novel insights on alignment, robustness, and interpretability.

• Ship what you publish: Turn research prototypes into evaluation libraries and client‑ready benchmarks adopted in production engagements.

• Strengthen our services: Provide deep‑dive analyses and red‑teaming support that raise the bar on every TaskUs delivery project.

What you’ll do:

• Design and run experiments on jailbreak detection, prompt‑safety evaluation, bias / toxicity measurement, and drift monitoring.

• Prototype safety interventions in Python using PyTorch / JAX / TensorFlow and libraries like LangChain or Tracr.

• Collaborate with your teams to refine hypotheses, share results, and prepare submissions to venues such as AIES, FAccT, or NeurIPS workshops.

• Support client projects by generating custom test suites, red‑team prompts, or interpretability reports under tight timelines.

• Document findings in clear technical reports, internal wikis, and client‑facing readouts.

• Keep a pulse on daily advances in AI safety, summarising relevant papers and open‑source releases for the broader team.

Experiences you’ll bring:

• 3–5 years of hands‑on ML or AI research / engineering experience, including 1+ year focused on safety, robustness, or trustworthy AI.

• MS in Computer Science, Machine Learning, or a related field or equivalent industry research background.

• Demonstrated work with large language models: fine‑tuning, evaluation, or safety assessments.

• At least one peer‑reviewed paper, preprint, or substantial internal research project.

• Proficiency in Python and one deep‑learning framework (PyTorch, TensorFlow, or JAX).

Core skills you'll need:

• Experimental design & analysis - hypothesis framing, statistics, and reproducibility.

• Prompt & policy evaluation, adversarial testing, and basic mechanistic interpretability.

• Rapid prototyping - comfortable iterating quickly to test new ideas.

• Clear communication - concise technical writing and slide decks for both researchers and clients.

• Collaboration & agility - willingness to jump onto a client request or pivot research focus as priorities shift.

Nice to have: Exposure to privacy‑preserving ML, or familiarity with NIST RMF / EU AI Act requirements.

Why you’ll love this role:

• Remote-first flexibility: Work where you’re most productive!

• Dual track: Publish high‑quality research and see it deployed in weeks, not years.

• Mentorship & growth: Work closely with a great team and expand your impact as the team grows.

• Fast lane to innovation: Daily access to frontier models and real client problems most researchers never see.

• Mission first: Help shape AI systems that are not only powerful but also safe and trustworthy.

Ready to put your research in the hands of real users?
Apply today and help build the future of trusted AI.

How We Partner To Protect You: TaskUs will neither solicit money from you during your application process nor require any form of payment in order to proceed with your application. Kindly ensure that you are always in communication with only authorized recruiters of TaskUs.

DEI: In TaskUs we believe that innovation and higher performance are brought by people from all walks of life. We welcome applicants of different backgrounds, demographics, and circumstances. Inclusive and equitable practices are our responsibility as a business. TaskUs is committed to providing equal access to opportunities. If you need reasonable accommodations in any part of the hiring process, please let us know.

We invite you to explore all TaskUs career opportunities and apply through the provided URL https://www.taskus.com/careers/.