Ubisoft Bordeaux

About the Company

Inaugurated in 2017 by a handful of Ubisoft veterans, the Ubisoft Bordeaux studio now has more than 420 talents from 20 different nationalities. The teams are currently working on some of Ubisoft's biggest AAAA games: as lead on Assassin's Creed Mirage, initiated and directed by the Bordeaux studio's creative and production teams, and co-dev on Beyond Good & Evil 2. In addition to these teams, the studio has developed a Tech branch, which works on all of Ubisoft's Online Services, as well as on the Anvil game engine. The studio has also created an R&D centre, La Forge, which brings Ubisoft engineers and researchers to work together on prototypes that will be used in game production, particularly around AI issues. Finally, the studio has an innovation department with a game in progress, not yet announced... Ubisoft Bordeaux will soon reunite all its teams in the Bastide-Niel eco-neighbourhood, in order to develop Ubisoft's best gaming experiences.

Listed Jobs

Company Name

Ubisoft Bordeaux

Job Title

[Research Internship] - Improving Non-Player Character Decision Making with ML (F/M/NB)

Job Description

**Job Title:** Research Internship: Improving Non-Player Character Decision Making with ML **Role Summary** Develop and implement deep learning methods to enhance non-player character (NPC) decision-making in video games, focusing on data-scarce scenarios. Integrate classical game AI with modern ML techniques to create realistic, responsive NPCs for applications such as combat AI, navigation, or adaptive gameplay strategies. **Expectations** - Focus on bridging classic game AI (e.g., behavior trees) with deep reinforcement learning techniques. - Collaborate with production teams to align research with game design and technical constraints. - Prioritize computational feasibility, realism, and scalability in NPC behavior models. **Key Responsibilities** - Research and evaluate state-of-the-art deep learning and reinforcement learning methods for decision-making. - Design hybrid architectures combining traditional AI with neural networks (e.g., residual learning, decision modulation). - Prototype and test solutions in 3D game environments. - Analyze performance metrics: realism, responsiveness, generalization, and computational efficiency. - Align prototypes with production requirements and gameplay scenarios. **Required Skills** - Proficiency in Python and deep learning frameworks (PyTorch, TensorFlow). - Strong understanding of machine learning, decision-making models, and reinforcement learning. - Knowledge of classical AI techniques (behavior trees, pathfinding, finite-state machines). - Experience with algorithms, optimization, probability, and linear algebra. **Required Education & Certifications** - Final year of an engineering degree or pursuing a research master’s in computer science or a related field. - No certifications required.

Bordeaux, France

Hybrid

21-10-2025

Company Name

Ubisoft Bordeaux

Job Title

[Research Internship] - Crafting NPCs & Bots behaviors with LLM/VLM - (F/M/NB)

Job Description

Job Title: Research Internship – NPCs & Bots Behaviors with LLM/VLM Role Summary: Conduct research on integrating Large Language Models (LLM) and Visual Language Models (VLM) with reinforcement learning (RL) or evolutionary strategies to enhance NPC and bot intelligence in video games. Work within a research lab providing realistic 3D game environments, datasets, and GPU resources. Collaborate with academic partners and production teams to identify viable use‑cases and develop prototype solutions. Expectations: • Follow a structured research plan over a 6‑month period. • Produce scientific experiments, performance analyses, prototype code, and documentation. • Communicate results to multidisciplinary teams in English. • Maintain rigorous version control, reproducibility, and code quality standards. Key Responsibilities: 1. Review literature on LLM/VLM applications in gaming and RL–based model grounding. 2. Design and implement experiments to embed LLM/VLM agents in interactive game environments. 3. Optimize learning pipelines (e.g., RL agents, evolutionary algorithms) for realistic NPC behavior. 4. Prototype scalable solutions using provided GPU clusters and game engines. 5. Analyze simulation data, evaluate performance gains, and iterate on model architectures. 6. Prepare technical reports, visualizations, and short presentations for internal review. 7. Contribute to collaborative discussions with production teams to define realistic use‑cases. Required Skills: - Proficiency in Python and modern ML frameworks (PyTorch, TensorFlow, or equivalents). - Experience with large‑scale transformer models (LLM/VLM) and fine‑tuning techniques. - Knowledge of reinforcement learning algorithms (policy gradients, Q‑learning, etc.) and evolutionary strategies. - Strong analytical skills: data preprocessing, statistical evaluation, performance metric design. - Ability to work independently while actively collaborating across interdisciplinary teams. - Excellent written and oral communication in English. Required Education & Certifications: • Final‑year engineering student or current master’s candidate in computer science, artificial intelligence, machine learning, or related field. • Solid foundation in computer science fundamentals and applied mathematics (probability, linear algebra, calculus). • No specific certifications required.

Bordeaux, France

Hybrid

21-10-2025

Company Name

Ubisoft Bordeaux

Job Title

[Internship] - R&D Developer Assistant - Neural Rendering for Real-Time Graphics (F/M/NB)

Job Description

Job Title: Internship – R&D Developer Assistant (Neural Rendering for Real‑Time Graphics) Role Summary: A 6‑month internship focused on researching and prototyping neural rendering techniques for real‑time graphics. The intern will survey the latest literature, identify limitations in current methods, and develop experimental GPU‑based prototypes to inform future integration into a real‑time rendering engine. Expectations: - Participate in weekly R&D meetings and presentations. - Manage end‑to‑end prototype development, from research to implementation. - Deliver functional code, performance benchmarks, and technical documentation. - Communicate findings clearly to technical directors and peers. Key Responsibilities: 1. Review and summarize cutting‑edge papers on neural appearance models, neural materials, and neural compression. 2. Analyze performance bottlenecks of neural rendering pipelines on modern GPUs. 3. Implement prototype modules (neural textures, neural lighting, compression) using C++, CUDA, and fragment/compute shaders. 4. Optimize memory usage and compute throughput to meet real‑time constraints. 5. Integrate prototypes into a test rendering pipeline and validate visual quality. 6. Document algorithms, experimental results, and best‑practice guidelines. 7. Present progress reports and demo sessions to the R&D team and artists. Required Skills: - Strong foundation in machine learning (network design, training, inference). - Proficiency in linear algebra, numerical optimization, and performance profiling. - Deep understanding of computer‑graphics fundamentals: rendering pipeline, shading, texture mapping, and GPU architecture. - Programming experience in C++, CUDA, GLSL/HLSL, or similar shading languages. - Familiarity with version control (Git) and code review practices. - Excellent analytical reasoning and self‑directed research ability. - Effective verbal and written communication in English. Required Education & Certifications: - Current second‑year master’s student or final‑year engineering student in Computer Science, Computer Graphics, Machine Learning, or a related discipline. - Completed coursework in machine learning, linear algebra, optimization, and computer graphics fundamentals. - No additional certifications required.

Bordeaux, France

On site

17-11-2025

Company Name

Ubisoft Bordeaux

Job Title

[Internship] R&D Developer Assistant - Real-Time Controllable Ocean Synthesis (F/M/NB)

Job Description

**Job title:** Internship – R&D Developer Assistant (Real-Time Controllable Ocean Synthesis) **Role Summary:** Provide research support for developing real-time ocean synthesis techniques. Focus on learning latent control spaces with neural embeddings, integrating neural BRDF models for water rendering, and creating demonstrators for real-time applications. **Expectations:** - Conduct literature reviews on ocean synthesis, latent space controllability, and neural rendering. - Identify gaps and evaluate existing methods. - Implement concise proof‑of‑concept prototypes. - Collaborate with researchers and engineers to discuss feasibility and results. **Key Responsibilities:** - Survey and document state-of-the-art research on procedural ocean synthesis and neural materials. - Design and train neural embedding models using offline simulation data. - Store embeddings in texture formats suitable for real-time access. - Extend render pipelines to incorporate neural BRDF modeling for water effects. - Build small real-time demonstrators to validate concepts. - Prepare technical summaries and communicate findings in English. **Required Skills:** - Strong foundation in machine learning, linear algebra, and optimization. - Knowledge of computer graphics fundamentals (rendering pipelines, shading, texture mapping). - Experience with neural rendering or neural material frameworks (e.g., block‑compressed features). - Programming in C++ or Python; familiarity with graphics APIs (OpenGL, DirectX, Vulkan) or game engine pipelines. - Analytical and problem‑solving abilities. - Effective written and verbal communication in English. **Required Education & Certifications:** - Current master’s student, second year of a research‑oriented program. - Coursework or experience in machine learning, computer graphics, or related fields. ---

Bordeaux, France

On site

17-11-2025