SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Spectrum Labs (trading as Alice) is seeking a Principal AI Research Scientist to lead hands-on research in frontier AI models, benchmarking, and evaluation. This is a highly technical, individual contributor role focused on driving fast-paced research cycles from hypothesis to validated findings.
You will conduct research in frontier AI models, reasoning systems, agents, and multimodal systems. Key responsibilities include designing and implementing AI benchmarks and evaluation frameworks, executing short research cycles with rapid iteration from prototypes to validated results, identifying model limitations and evaluation gaps, and working directly with customers and technical stakeholders to translate complex challenges into actionable research outcomes. You'll collaborate with engineering and product teams to operationalize research findings into practical solutions.
Alice is a trust, safety, and security company focused on safeguarding AI systems across the entire lifecycle. The company provides model hardening evaluations, pre-deployment red-teaming, runtime guardrails, and drift detection for frontier model labs, enterprises, and user-generated content platforms. Alice serves over 3 billion users across major AI and tech platforms and is recognized as a global leader in online safety and AI security.
REQUIREMENTS:
- PhD or MSc in Computer Science, AI, Machine Learning, or related field with strong relevant research experience
- Strong publication record at leading AI/ML conferences (NeurIPS, ICML, ICLR, ACL, or equivalent)
- Exceptional hands-on technical capabilities: advanced Python, PyTorch, and experience building and evaluating complex AI systems
- Proven ability to independently design, implement, and execute research experiments within short development cycles
- Deep understanding of LLMs, foundation models, benchmarking, and evaluation methodologies
- Strong analytical and communication skills, with ability to operate effectively in fast-paced, customer-driven research environment
PREFERRED QUALIFICATIONS:
- Experience at frontier AI labs (OpenAI, Anthropic, Google DeepMind)
- AI for Cybersecurity background: autonomous security agents, vulnerability research, penetration-testing automation, security operations
- Security of AI expertise: LLM red teaming, adversarial machine learning, prompt injection, agentic AI security
- Familiarity with cybersecurity tools and environments (CTFs, sandboxed exploitation, static/dynamic code analysis, security testing frameworks)
- Experience developing cybersecurity benchmarks or evaluating frontier models on complex security tasks
