Senior Software Engineer – AI Agent Platform (W/M/X)

At Ubisoft, we are building a cloud platform that turns meaningful jobs into dependable agent work. This platform provides the foundations needed to build, run, and operate AI agents securely and at scale. It coordinates agent execution, manages reusable agent definitions and resources, maintains durable sessions, provides isolated execution environments, and connects agents to models, tools, knowledge, and internal services. Our goal is not simply to demonstrate what AI agents can do. We want to make them reliable enough for real production use: observable, secure, scalable, governable, and accessible through a range of applications and services. We are seeking an experienced Senior Software Engineer to help design and build these platform services, orchestration engine, and agent runtimes. You will work on distributed back-end systems that coordinate long-running work, provision isolated execution environments, stream results, manage failures, and serve multiple products and teams. This is an opportunity to work at the intersection of distributed systems, cloud infrastructure, developer platforms, and applied AI—with significant scope for technical ownership and impact. Responsibilities As a Senior Software Engineer on the Kobo platform, you will help turn rapidly evolving agent capabilities into dependable production systems. Design and build platform services Design, implement, and operate scalable back-end services for agent orchestration, execution, sessions, resource management, and platform control. Define clear service responsibilities, domain boundaries, and APIs that allow components and consumers to evolve independently. Build dependable agent runtimes Contribute to the systems that provision and operate isolated environments in which agents perform their work. Address challenges such as workload isolation, resource management, lifecycle control, streaming, cancellation, failure recovery, and communication with models, tools, knowledge systems, and internal services. Develop orchestration capabilities Build systems that coordinate agent runs across services and runtimes, carrying the appropriate identity, context, configuration, and resources throughout the execution lifecycle. Design for asynchronous and long-running work, partial failures, retries, idempotency, and durable state where required. Engineer for production Ensure that platform services are reliable, observable, secure, and cost-effective. Establish meaningful metrics, traces, logs, health signals, and operational tools. Diagnose failures across service boundaries and improve the platform based on production evidence. Build cloud-native infrastructure Design and operate services on AWS using containerized and orchestrated environments. Work with technologies such as Kubernetes, container runtimes, virtualized execution environments, managed cloud services, and event-driven infrastructure. Protect platform boundaries Treat identity, authorization, secret management, tenant isolation, and untrusted inputs as fundamental design concerns. Work closely with infrastructure and security specialists to ensure that agent workloads operate within explicit and enforceable boundaries. Collaborate across teams Partner with product engineers, infrastructure teams, security specialists, and teams building applications on Kobo. Translate product needs into reusable platform capabilities while avoiding application-specific coupling in the platform. Provide technical leadership Lead technical design discussions, make and document pragmatic architecture decisions, review code, and mentor other engineers. Help improve engineering practices while remaining actively involved in implementation, testing, deployment, and production operations. Evaluate emerging technology pragmatically Follow developments in agent frameworks, runtime isolation, model access, and cloud infrastructure. Evaluate new technology based on concrete platform needs rather than novelty, and help determine when to adopt, adapt, or avoid it. We are looking for a seasoned software engineer who enjoys building foundational systems and taking responsibility for how they behave in production. Must-Have Qualifications Professional experience : Strong practical software engineering experience, with substantial experience building back-end, infrastructure, or platform systems in a production environment. Programming skills : Strong proficiency in at least one production back-end language, such as Python, Go, Rust, Java, C++, or TypeScript, and the ability to work effectively across an evolving technology stack. Distributed systems : A strong understanding of distributed-system concepts, including consistency, concurrency, asynchronous processing, retries, idempotency, partial failure, and service-to-service communication. System and API design : Experience designing maintainable services and APIs with clear ownership, stable contracts, and appropriate boundaries between transient coordination and durable state. Cloud and container platforms : Hands-on experience with a major cloud platform—ideally AWS—and container technologies such as Docker and Kubernetes. Production engineering : Experience operating services in production, including monitoring, observability, incident diagnosis, performance analysis, capacity planning, and reliability improvements. Security mindset : An understanding of authentication, authorization, secret management, workload isolation, and the risks involved in executing user- or agent-directed workloads. Problem-solving : A demonstrated ability to navigate ambiguous, complex technical problems and turn them into pragmatic, maintainable solutions. Collaboration: Strong communication skills and experience working across product, platform, infrastructure, and security teams. Nice-to-Have Qualifications Experience building workflow engines, orchestration systems, schedulers, developer platforms, or execution runtimes. Experience with sandboxing, container isolation, virtual machines, microVMs, or technologies such as Firecracker. Familiarity with AI agents, agent loops, tool calling, model gateways, skills, MCP servers, or related integration patterns. Experience with streaming protocols and real-time communication, including Server-Sent Events, WebSockets, or similar technologies. Experience designing multi-tenant systems with strong authorization and isolation requirements. Familiarity with event-driven systems, queues, and long-running asynchronous workflows. Prior experience training machine-learning models is not required. We value experience making complex systems reliable in production more than expertise in a specific model or framework. Growth and Impact Joining the Kobo team offers an opportunity to help define how AI agents are built, executed, and operated across Ubisoft. You will have the opportunity to: Shape the architecture and engineering principles of an emerging AI agent platform. Work on foundational capabilities used by multiple products and teams. Solve challenging problems involving distributed systems, runtime isolation, security, and cloud-scale operation. Influence technical strategy while remaining closely involved in implementation. Learn from real production workloads and use that evidence to evolve the platform. Take increasing ownership of major platform domains and mentor other engineers. Help establish what dependable, production-grade agent execution means in practice. Ubisoft's perks 💰 Profit Sharing, yearly company saving plan. 25 paid time off + 12 additional paid days off. 50% of your transportation pass is paid by the company, lunch vouchers (9€/day), healthcare for you and your family, and lots of Ubisoft additional perks. 👶 Maternity leaves of 20 weeks, paternity/co-parental leaves of 7 weeks. 📍 Our office is located in Saint Mandé, (Metro line 1, Saint Mandé station). Gym available in the building. Information about Ubisoft Ubisoft offers the same job opportunities to all, without any distinction of gender, ethnicity, religion, sexual orientation, social status, disability, or age. Ubisoft ensures the development of an inclusive work environment which mirrors the diversity of our gamers’ community.