← All roles

Staff Applied AI Researcher - Agentic Reasoning Systems (Dublin, CA)

Articul8· Applied Research
Apply on company site
Dublin, CA (HQ)Full-timePosted 5mo ago

About the role

Staff Applied AI Researcher to define runtime reasoning and trustworthy autonomous decisioning for Articul8's ModelMesh-powered platform. Lead research across runtime intelligence, agent coordination, model/tool routing, verification, observability-aware execution, and enterprise-grade evaluation methods.

Responsibilities

  • Set technical direction for agentic reasoning systems and runtime intelligence across ModelMesh™ — define the orchestration strategies, decision policies, verification approaches, and runtime quality standards that determine how massively parallel agent systems reason, coordinate, and self-correct in production
  • Architect the infrastructure for researcher augmentation at scale — design the agentic platforms and orchestration primitives that enable every researcher and engineer at Articul8 to deploy fleets of AI agents for experimentation, evaluation, and production integration — multiplying the depth, breadth, and velocity of the entire organization
  • Go deep: advance the science of autonomous reasoning — design, train, and refine the learned components behind runtime decisioning (routing models, verification models, confidence estimators, reward models, policy selectors), using massively parallel agent-driven experiment pipelines to explore architectural and algorithmic frontiers exhaustively
  • Go broad: unify perception, retrieval, reasoning, and action — build repeatable methodology for composing domain-specific models, data perception systems, knowledge graphs, retrieval layers, and external tools into coherent agentic workflows, delegating integration testing and cross-modal benchmarking to parallel agent systems so you can reason across the full stack simultaneously
  • Drive research on agent reliability for regulated environments — lead failure detection, self-checking, verification workflows, compounding error analysis, and auditable autonomous behavior research, using agent-orchestrated stress testing and red-teaming at scales that manual evaluation cannot reach
  • Define evaluation methodology for runtime intelligence — establish how task success, decision quality, robustness, traceability, and failure recovery are measured under realistic enterprise conditions, building agentic evaluation harnesses that run continuously and surface regressions before they reach customers
  • Influence platform-level architecture — shape decisions on model routing, tool use, observability, governance, access control, and interoperability with external agent ecosystems, ensuring the platform is designed for humans and agents to amplify each other
  • Mentor researchers across levels in the agentic paradigm — raise the bar on technical judgment, experimental rigor, and agent-augmented research practice; contribute to hiring researchers who are driven to maximize their human potential
  • Maintain hands-on research impact — sustain a meaningful personal research contribution through technical work, publications, patents, and externally visible output, modeling what it looks like to be a deeply technical leader who uses agentic systems to go deeper and faster than ever before

Qualifications

  • Education: PhD or MSc in Computer Science, Machine Learning, AI, Robotics, or a related field.
  • Experience: 8+ years in AI/ML research with demonstrated impact on production systems, including 3+ years building LLM-based or autonomous AI systems.
  • Reasoning and orchestration: Deep hands-on experience in at least two of: multi-agent coordination, planning under uncertainty, sequential decision-making, probabilistic inference, model routing, or tool-using agent systems.
  • Evaluation of autonomous systems: Designed evaluation frameworks for systems where correctness is not binary — measuring decision quality, reliability under distribution shift, compounding error rates, and failure recovery in production-like conditions.
  • Systems at scale: Designed and operated research systems that integrate multiple models, data sources, and control mechanisms in production or near-production settings.
  • Software engineering: Proficient in Python with strong software architecture instincts.
  • Technical leadership: Set technical direction for a research area, mentored researchers, and influenced quality standards beyond the immediate team.

Skills

Required:PhD or MSc in relevant fieldAI/ML research with production impactMulti-agent coordination / planning under uncertainty / sequential decision-makingEvaluation frameworks for autonomous systemsPython programming and software architectureTechnical leadership and mentoring
Last updated Aug 5, 2026. Interested? Apply on the company site.
Articul8 logo

About Articul8

The GenAI platform that simply works.

The Domain-Specific GenAI Platform, Purpose-Built for Your Data and Mission

GenAIArtificial IntelligenceAI Platform
Company profileWebsite

More open roles at Articul8