Georgi Kardzhaliyski
AI Safety & Product Leader
π’ Now interviewingFormer AI Red Teamer & Prompt Engineer at Meta (Oct 2025 - Mar 2026), where I improved Llama model safety through adversarial testing. Now interviewing for senior roles in AI safety, product, or AI strategy. 10+ years driving quantified impact across the U.S. and Europe; NYC-based U.S. Permanent Resident with immediate availability.

About Me
Georgi has worn many hats throughout his career, working with CEOs, engineering teams, designers, product and project managers across startups, mid-sized and large corporations, and NGOs. His international background helps him navigate the nuances of both U.S. and European business cultures. Georgi thrives in start-up environments, even when chaotic, where he can make an impact alongside ambitious international colleagues who challenge his thinking.
Georgi holds a BS in Management Information Systems (MIS) and Economics from Worcester Polytechnic Institute (WPI), an MBA from Cambridge College, and specialized graduate studies from the Copenhagen Business School (CBS) and WU Wien (Vienna University of Economics and Business).
While at WPI, Georgi and his team published their MIS capstone research on how mobile user experience (UX) drives business value at international Human-Computer Interaction (HCI) conferences. Beyond academic work, he collaborated with the Bulgarian-American Fulbright Commission to publish and distribute a collection of stories to high schools throughout Bulgaria.
When Georgi is not working, you can find him training for his next marathon. He is pursuing running all the Abbott World Marathon Majors, a series of the world's 7 most prestigious marathons. So far, Georgi's completed 3/7: Chicago (2019), Berlin (2021), and New York City (2022), plus six other marathons along the way. Still to come: Sydney, Tokyo, Boston, and London.
Featured Work
AI Red Teamer & Prompt Engineer - Product & Research
Oct 2025 - Mar 2026Meta Platforms, Inc.
Improved Llama model safety at Meta Platforms by conducting adversarial testing and red-teaming to uncover jailbreak vulnerabilities and policy-violating outputs. Analyzed and categorized model responses for safety and policy compliance using Meta's taxonomy.
Tech Startup Product Leadership
2013-2022Mitrend
Led 5 product launches at Mitrend (Boston-based tech startup), managing cross-continental engineering and design teams. Conducted 20+ UX studies and achieved 20% increase in product-market fit.
AI Strategy & Business Transformation
2024-2025Advance Valuations
Led comprehensive AI adoption initiative at Advance Valuations, identifying 20+ automation opportunities and training 50+ employees. Designed and implemented AI solutions that automated routine tasks across core business functions.
Earlier Career & Consulting
Strategic work for Fortune 500 brands across Europe and the U.S.
Get In Touch
Formerly AI Safety & Red-Teaming at Meta (Oct 2025βMar 2026). Now actively interviewing for senior roles in AI safety, product, and AI strategy.
π‘ Want to know more? Ask my AI assistant about my skills, experience, or projects!