AI Security Researcher
Grammarly is excited to offer a remote-first hybrid working model. Team members work primarily remotely in the United States, Canada, Ukraine, Germany, or Poland. Certain roles have specific location requirements to facilitate collaboration at a particular Grammarly hub.
All roles have an in-person component: Conditions permitting, teams meet 2–4 weeks every quarter at one of Grammarly’s hubs in San Francisco, Kyiv, New York, Vancouver, and Berlin, or in a workspace in Kraków. This flexible approach gives team members the best of both worlds: plenty of focus time along with in-person collaboration that fosters trust and unlocks creativity.
Grammarly team members in this role must be based in the United States or Canada, and they must be able to collaborate in person 3 weeks per quarter, traveling if necessary to the hub(s) where the team is based.
Grammarly is the world’s leading AI writing assistance company trusted by over 30 million people and 70,000 professional teams every day. From instantly creating a first draft to perfecting every message, Grammarly’s product offerings help people at 96% of the Fortune 500 get their point across—and get results. Grammarly has been profitable for over a decade because we’ve stayed true to our values and built an enterprise-grade product that’s secure, reliable, and helps people do their best work—without selling their data. We’re proud to be one of Inc.’s best workplaces, a Glassdoor Best Place to Work, one of TIME’s 100 Most Influential Companies, and one of Fast Company’s Most Innovative Companies in AI.
To achieve our ambitious goals, we’re looking for a highly skilled and motivated AI Security Researcher to join Grammarly Security. This is a unique opportunity for a hands-on technical lead to own the vision, technical direction, and roadmap for AI security at the company globally. Your findings and contributions will directly impact the development of robust AI technologies and safeguard all Grammarly product offerings and services against emerging security and responsible AI risks. User trust is at the heart of everything we do.
Grammarly’s security researchers have the freedom to develop practical solutions that, in turn, influence our security culture. The engineering landscape at Grammarly is complex as we create innovative product offerings, expand to new platforms, and scale underlying cloud systems—all while strengthening and expanding our global team. Read more about our stack or hear from our team on our technical blog.
As an AI Security Researcher, you will:
- Explore cutting-edge techniques, develop innovative methodologies, and advance the practice of secure and safe AI development at Grammarly.
- Conduct in-depth research, analyze AI systems, and develop novel methodologies and techniques to proactively detect and mitigate security risks in AI systems, including adversarial attacks, data poisoning, model evasion, and others.
- Evaluate first- and third-party AI models used in Grammarly product offerings to identify and mitigate security risks that could negatively impact our customers.
- Investigate the impact of emerging technologies, such as large language models, deep learning, and reinforcement learning, on AI security and safety and develop countermeasures accordingly.
- Collaborate with cross-functional teams, including AI developers, data scientists, linguists, responsible AI researchers, and ML engineers, to integrate security improvements and best practices into Grammarly's AI development lifecycle, from low-hanging fruit to long-term investments.
- Participate in internal research discussions, brainstorming sessions, and technical reviews to ensure rigorous scientific methodology and foster innovation.
- Stay abreast of AI security and safety research advancements, attend conferences, and actively contribute to the security community through publications and presentations.
We’re looking for someone who
- Embodies our EAGER values—is ethical, adaptable, gritty, empathetic, and remarkable.
- Is inspired by our MOVE principles, which are the blueprint for how things get done at Grammarly: move fast and learn faster, obsess about creating customer value, value impact over activity, and embrace healthy disagreement rooted in trust.
- Is able to collaborate in person 3 weeks per quarter, traveling if necessary to the hub where the team is based.
- Has a Ph.D. in Computer Science, Artificial Intelligence, Cybersecurity, or a related field with a focus on AI security and safety research.
- Has a strong background and expertise in machine learning, deep learning, and neural networks.
- Has a solid understanding of security fundamentals, such as threat modeling, cryptographic protocols, and vulnerabilities.
- Is proficient in programming languages such as Python or Java and has experience with relevant AI frameworks and tools.
- Has a proven track record of conducting research, demonstrated through publications, including at top-tier conferences or journals.
Support for you, professionally and personally
- Professional growth: We believe that autonomy and trust are key to empowering our team members to do their best, most innovative work in a way that aligns with their interests, talents, and well-being. We support professional development and advancement with training, coaching, and regular feedback.
- A connected team: Grammarly builds a product that helps people connect, and we apply this mindset to our own team. Our remote-first hybrid model enables a highly collaborative culture supported by our EAGER (ethical, adaptable, gritty, empathetic, and remarkable) values. We work to foster belonging among team members in a variety of ways. This includes our employee resource groups, Grammarly Circles, which promote connection among those with shared identities, such as BIPOC and LGBTQIA+ team members, women, and parents. We also celebrate our colleagues and accomplishments with global, local, and team-specific programs.
Compensation and benefits
Grammarly offers all team members competitive pay along with a benefits package encompassing the following and more:
- Excellent health care (including a wide range of medical, dental, vision, mental health, and fertility benefits)
- Disability and life insurance options
- 401(k) and RRSP matching
- Paid parental leave
- Twenty days of paid time off per year, eleven days of paid holidays per year, and unlimited sick days
- Home office stipends
- Caregiver and pet care stipends
- Wellness stipends
- Admission discounts
- Learning and development opportunities
Grammarly takes a market-based approach to compensation, which means base pay may vary depending on your location. Our US and Canada locations are categorized into compensation zones based on each geographic region’s cost of labor index. For more information about our compensation zones and locations where we currently support employment, please refer to this page. If a location of interest is not listed, please speak with a recruiter for additional information.
Base pay may vary considerably depending on job-related knowledge, skills, and experience. The expected salary ranges for this position are outlined below by compensation zone and may be modified in the future.
We encourage you to apply
At Grammarly, we value our differences, and we encourage all—especially those whose identities are traditionally underrepresented in tech organizations—to apply. We do not discriminate on the basis of race, religion, color, gender expression or identity, sexual orientation, ancestry, national origin, citizenship, age, marital status, veteran status, disability status, political belief, or any other characteristic protected by law. Grammarly is an equal opportunity employer and a participant in the US federal E-Verify program (US). We also abide by the Employment Equity Act (Canada).
Please note that EEOC is optional and specific to US-based candidates.
All team members meeting in person for official Grammarly business or working from a hub location are strongly encouraged to be vaccinated against COVID-19.