AI Researcher

Role Overview

This opportunity is shared on behalf of a partner employer that oversees the full hiring process. The company is seeking an AI Researcher in Ireland to help advance agentic AI systems that learn from real interaction data at scale. The work centers on turning multimodal information such as text, audio, logs, and structured workflows into stronger reasoning and decision-making models.

The position combines hands-on research with product impact. You will work at the intersection of scientific experimentation and real-world deployment, with research outcomes expected to improve live user-facing agents. Success in this role requires both rigorous research thinking and the ability to build practical systems. It is well suited to someone comfortable with uncertainty, complex problem solving, and shaping the direction of multimodal and agentic AI.

Responsibilities

• Carry out advanced research on agentic AI systems trained with real-world interaction data, with an emphasis on better reasoning, planning, and tool usage.
• Build and test learning approaches such as RAG, fine-tuning, RLHF, DPO, and GRPO to raise large-model performance.
• Create multimodal representation-learning methods, including shared embedding spaces for text, audio, logs, and structured data.
• Enhance speech and audio intelligence capabilities, including STT, ASR, and audio-based learning pipelines.
• Establish evaluation methods to assess agent performance in practical and domain-specific settings.
• Convert complex behavioral and interaction signals into structured training objectives for large-scale models.
• Partner with engineering and product teams to move research into production and improve it using live feedback.

Requirements

• Doctoral degree in Computer Science, Machine Learning, AI, Electrical Engineering, or a closely related discipline.
• At least 5 years of experience in applied AI research or ML systems with production impact.
• Deep knowledge of large-scale machine learning, LLMs, or multimodal AI systems.
• Practical experience with RAG, LLM fine-tuning, and reinforcement learning techniques such as RLHF, DPO, or GRPO.
• Strong foundation in representation learning, embeddings, and shared multimodal spaces.
• Experience working with speech and audio models, including STT, ASR, or audio signal processing.
• Strong programming ability in Python and familiarity with modern ML tools such as PyTorch and Hugging Face.
• Ability to design evaluation frameworks for LLMs or agent-based systems.
• Comfort working through ambiguous real-world challenges and forming research hypotheses from incomplete signals.
• Excellent written and spoken English communication skills.

Perks and Benefits

• Fully remote work arrangement within a global AI research organization.
• Chance to contribute to advanced agentic and multimodal AI products.
• Research work with direct influence on production systems.
• Close collaboration with experienced engineering and product teams.
• Strong ownership of research direction and technical strategy.
• Access to large proprietary datasets and AI infrastructure.
• Competitive compensation based on experience and expertise.

Additional Information

This role is part of a partner-company hiring process. Applications are reviewed by the partner employer, while the listing platform may assist with matching and routing candidates. Final interview stages and hiring decisions are managed by the employer’s internal team.

Data Privacy and Hiring Process

By applying, candidates agree that their personal data may be used to assess suitability for the role and shared with the hiring employer where relevant. The processing is based on legitimate interest and pre-contractual steps under applicable data protection laws, including GDPR. Applicants may request access, correction, deletion, or objection at any time.

Some parts of the screening workflow may use AI tools to support resume review, response analysis, and consistency checks. These tools assist the recruitment team and do not replace human judgment. Final hiring outcomes are determined by people. Applicants can contact the employer for more details about data handling.