Senior Researcher, Explainable Reasoning - Unikie Oy - Remote - Global

We are looking for a Senior Researcher, Explainable Reasoning to our team. This role focuses on advancing cutting-edge AI systems that can reason over complex multimodal data (text, images, video, and knowledge sources) while remaining interpretable, safe, and aligned with human values. The researcher will design novel AI architectures, improve reasoning capabilities in LLMs and agentic systems, and develop explainability and safety techniques that make AI behavior more transparent, reliable, and resistant to misuse. Main Responsibilities include developing advanced multimodal AI systems capable of reasoning across text, images, video, and cultural knowledge; researching and implementing explainable AI (XAI) and neural-symbolic reasoning techniques; improving transparency, interpretability, fairness, and safety of AI/LLM systems; designing alignment methods for LLMs and agentic AI workflows using techniques such as RLHF, RLAIF, Constitutional AI; reverse-engineering neural networks; building data attribution and interpretability tools; creating benchmarks and datasets for evaluating reasoning quality, reliability, and explainability; developing defenses against adversarial attacks, jailbreaks, and prompt injection threats; translating state-of-the-art AI research into scalable engineering solutions; and communicating technical findings and research insights to stakeholders.