Identify and define critical AI safety and security problems,establish technical success criteria.
Conduct direct research and experiments to build code,models,data,and systems validating hypotheses.
Advance new attack and defense methodologies,evaluation frameworks,and core Stinger and Starfort technologies.
What they want
Demonstrated deep technical expertise and representative achievement in ML/LLM research,AI safety/security,adversarial ML,offensive/defensive security,ML/Agent systems,or software/AI product engineering.
Experience structuring and executing undefined problems to meaningful completion independently.
Capability to implement solutions directly via code,models,experiments,or systems rather than ideation alone.
Nice to have
Priority awareness of company's most critical problems over stated job scope.
Capacity to develop deep expertise in one domain while rapidly learning new domains as needed.
Understanding that rapid experimentation and production-grade reliability require different standards.
Seoulstart's read
Visa tier (likely)
E-7 specialty / professional (likely)
Korean language
Business-level Korean required
English friendliness
Medium: English JD, mixed-language workplace
Context
AI safety startup with major VC backing (Samsung Ventures,Mirae Asset,etc.) and partnerships with Anthropic/OpenAI/Google/Microsoft; Korean-fluent communication expected for internal collaboration despite global investor base and international partnerships.
This is Seoulstart's analysis of the public job description, not the employer's stated policy. Verify visa sponsorship, language requirements, and remote allowances directly with the recruiter.