Role overview
As a Research Scientist/Engineer in Open-Endedness, you will develop the generative and adaptive intelligence at the heart of our virtual entities. Your work will focus on building systems capable of continuous novelty—AI that creates an endless stream of compelling behaviors, narratives, and interactions tailored to any audience.
As part of a small, focused team, you'll have significant autonomy and end-to-end ownership. You will work at the frontier of open-ended learning, foundation model agents, and methods for continual adaptation. You'll tackle questions like: How do we build characters that never run out of interesting things to do? How do we create systems that genuinely adapt to radically different users? How do we sustain novelty and surprise without sacrificing coherence or quality?
We embrace simple methods that scale. We're looking for someone excited to build robust infrastructure, develop rigorous evaluation methods, and push the boundaries of what open-ended AI systems can achieve in interactive contexts.
At Iconic, you'll join a team that uniquely blends cutting-edge AI research with AAA game development experience. The successful candidate will work alongside not just fellow researchers, but also designers, artists, and animators. This is a unique opportunity to shape the generative core of AI characters in the interactive experiences of tomorrow.
Responsibilities
- Research and develop open-ended systems that generate continuous streams of diverse, engaging behaviors and content
- Develop foundation model agent architectures for adaptive, context-aware reasoning and decision-making
- Design systems that enable virtual entities to adapt their behavior to different audiences, contexts, and situations in real-time
- Create methods for continual learning and adaptation in dynamic, unbounded settings
- Design evaluation and reward frameworks and metrics for novelty, diversity, engagement, and out-of-distribution generalization
- Collaborate closely with other researchers, artists, and animators to unlock new creative experiences
- Publish and present research at top-tier venues (NeurIPS, ICLR, ICML)
- Stay current with and contribute to the state of the art in open-ended learning and foundation model agents
Basic qualifications
- MSc or PhD in Computer Science, Machine Learning, Artificial Intelligence, or a related field (or equivalent industry experience)
- Strong foundation in deep reinforcement learning and/or foundation model agents
- Experience with open-ended learning approaches (e.g., evolutionary methods, quality-diversity, intrinsic motivation, population-based training, self-play)
- Experience prompting, evaluating, and fine-tuning LLMs, and/or building LLM-based agents
- Proficiency in Python and deep learning frameworks (PyTorch, JAX, or TensorFlow)
- Strong research and engineering skills: ability to build robust systems while advancing the state of the art
- Excellent collaboration and communication skills
Preferred qualifications
- Strong publication record at top-tier venues (NeurIPS, ICLR, ICML)
- Experience building training codebases for LLMs, RL agents, or open-ended methods in complex environments
- Expertise optimizing distributed training or inference systems
- Experience with procedural content generation or AI-driven narrative systems
- Background in artificial life, complex systems, or emergent behavior
- Familiarity with game engines (Unreal, Unity) or interactive systems
Benefits
- Competitive salary and equity compensation
- 25 days annual leave + bank holidays
- Private healthcare
- Based in London with hybrid work
- Inclusive & friendly company culture with socials and game breaks
About the company
Iconic Interactive is a seed-stage startup building AI that breathes life into virtual worlds. The future of entertainment is personal: entire universes shaped around each of us, where you are not watching a story but living at the center of one, shaping it. We're building every layer of intelligence these experiences need: characters that feel and convey meaning, narrators that weave your story, and world directors that act like an ever-present game master: adapting, orchestrating, surprising. We're a growing team tackling some of the most fascinating problems in AI: creating minds that inhabit and shape new worlds.
The Mission
Virtual characters today are fundamentally limited: they run out of things to say, repeat patterns, and fail to surprise. At Iconic, we're building something different: digital entities that resemble improv actors, with the capacity for endless creativity and adaptation. We're looking for an AI Scientist/Engineer to help us develop the systems that make our characters perpetually compelling: generating novel behaviors, adapting to diverse audiences, and sustaining engagement without feeling scripted or predictable.