Agentic Staff Engineer

Requirements

• Professional English proficiency is required (written and spoken)
• Extensive experience designing and operating large-scale distributed systems in production (reliability, performance, cost, security)
• Deep understanding of agentic patterns and their failure modes (tool/function calling, orchestration, memory, planning/routing, evaluation)
• Proven track record of technical leadership across multiple teams: setting standards, influencing roadmaps, and delivering platforms used by others
• Strong system design skills including multi-tenant architectures, permission models, data governance, and safe execution of actions
• Strong operational mindset: observability, production debugging, incident response, and driving durable corrective actions
• (Desirable) Experience building evaluation systems at scale (test suites, synthetic data, human review pipelines, automated regression, A/B testing)
• (Desirable) Strong understanding of privacy-by-design and compliance constraints for user/content data (PII handling, retention, access control)
• (Desirable) Experience with retrieval systems (vector search, hybrid retrieval, ranking) and cost-control patterns for LLM-heavy products
• (Desirable) Familiarity with event-driven/workflow patterns (queues, retries, idempotency, saga patterns) and platform engineering practices

What the job involves

• At Dailymotion, we’re building AI-powered products that rely on high-quality, well-governed data—from advertising optimization and measurement to trust & safety and internal decision-making tools
• We are looking for an Agentic Engineer to design and operate the data pipelines, datasets, and evaluation workflows that make our AI systems reliable, reproducible, and measurable in production
• In this role, you’ll work at the intersection of data engineering, ML/LLM product needs, and platform reliability, partnering closely with AI/ML engineers, product engineers, and analytics stakeholders
• You’ll help ensure that the data powering training, retrieval (RAG), evaluation, and monitoring is curated, traceable, and secure—so teams can ship AI features with confidence
• We offer challenging work with direct business impact, opportunities to learn and grow, and a collaborative culture that encourages every team member to bring their point of view to the table
• Own the technical vision and architecture for agentic systems: multi-step workflows that use tools, memory/state, planning, and human-in-the-loop controls
• Design and evolve the internal agent orchestration stack built on the conversation manager and Eino: tool registry, permissions, execution lifecycle, state management, and audit trails
• Define engineering standards for agent reliability and debuggability: traceability, deterministic replay where possible, evaluation harnesses, robust fallbacks, and safe degradation modes
• Lead cross-team initiatives spanning Product, Data, Infrastructure, Security/Privacy, and ML; align stakeholders and ensure delivery across organizational boundaries
• Set direction on platform choices and patterns: model routing on Vertex AI, caching strategies, prompt/config management, observability, and governance
• Drive platform adoption through shared libraries, reference implementations, documentation, and internal enablement
• Establish safety and compliance controls: data access governance, prompt-injection defenses, secrets management, policy enforcement, and secure tool execution
• Mentor senior engineers and raise the technical bar via architecture reviews, operational excellence, and pragmatic decision‑making
• Start Date: Asap
• Contract Type: Permanent