AI Safety Argumentation Research Engineer

This is a fully remote position, open to applicants in United States.

📋 Description

• Expand the ontologies and knowledge graph schemas that represent claims, evidence, argument structures, defeaters, and levels of confidence.

• Develop defeasible argumentation frameworks (e.g., ASPIC+, Dung-style, argumentation schemes) that encompass both logical structure and susceptibility to rebuttal.

• Manage and ensure quality control of LLM-driven population pipelines, incorporating cross-check scaffolds, provenance tracking, and human-in-the-loop curation.

• Design agent coordination patterns for multi-step research and population tasks, ensuring robust error management and graceful degradation.

• Pre-structure argument frameworks by identifying the strongest counterarguments, steel-manned objections, and known defeaters.

• Create export pipelines that convert structured argumentation into various communication formats suitable for different audiences and contexts.

• Stay informed about developments in AI safety, capabilities, and governance to recognize when updates to the graph are necessary, and to identify authoritative sources for further information.

• Collaborate with communications personnel and researchers to ensure that outputs meet real persuasive needs.

⛳️ Requirements

• Practical knowledge of formal or semi-formal argumentation theory (including abstract or structured argumentation, defeasible reasoning, dialectical models, or argumentation schemes).

• Experience in ontology engineering or knowledge graph development (using OWL/RDF, property graphs, or similar technologies).

• Hands-on experience with LLM agent systems: including agent coordination platforms, large-scale prompt engineering, and quality control processes for LLM outputs (such as adversarial probing, consistency checks, and calibration).

• Proficient in vibecoding: quickly prototyping and deploying LLM-assisted development in production-adjacent environments.

• Strong foundational knowledge in AI safety, AI governance, and current frontier-AI dynamics, with the ability to locate authoritative resources on any related sub-topic or expertise.

• Understanding of philosophical concepts relevant to evidence, such as defeaters, burden of proof, inference to the best explanation, and underdetermination.

• Competent coding skills; familiarity with graph databases or query languages.

• Experience in designing cross-check and verification frameworks for unreliable automated processes.

• Good judgment regarding when a claim is well-supported and when it requires hedging, additional substantiation, or withdrawal.

• Self-motivated with strong written communication abilities.

• Graduate-level education or equivalent experience in argumentation theory, computational argumentation, epistemology, or philosophy of science.

• Familiarity with computational argumentation tools such as AIF or Carneades.

• Proven experience in AI safety or governance, including publications, policy work, or significant contributions to the community.

• Background in argument mining, claim extraction, or stance detection.

• Experience with debate formats or structured deliberation techniques.

• Understanding of motivated reasoning, belief change, and cognitive biases as they relate to communication strategies.

• Contributions to open-source projects in any relevant domain.

🏝️ Benefits

• Competitive salary and performance-based incentives.

• Opportunities for professional development and continuous learning.

• Flexible work environment with remote work options.

• Collaborative and inclusive company culture.

• Comprehensive health and wellness benefits.