Hiring via MeritMerit connects world-class engineers and researchers with the most innovative startups and scale-ups. We're currently hiring aFounding Lead AI Scientistfor aventure-backed AI companythat's building the next generation of agentic systems for marketing automation and growth.About the CompanyThis early-stage, AI-first company is redefining how B2B marketing teams grow. Their autonomous platform replaces dashboards and manual workflows with intelligent agents that plan, act, and learn — spinning up campaigns, generating creative, pacing budgets, and self-correcting in live traffic.They've recently raised their seed round and are expanding rapidly across the US and Europe. This is a rare opportunity to define the scientific and architectural foundations of an AI company whereautonomy, safety, and scalecome together.The OpportunityAs theFounding Lead AI Scientist, you'll architect and deliver the next generation ofagentic AI systems. You'll design, train, and deploy autonomous agents that make and execute real-time growth decisions — operating across research, engineering, and product boundaries.You'll move fluidly from notebook to production, establishing experimentation frameworks, building evaluation harnesses, and translating cutting-edge AI research into tangible business outcomes.In your first 12 months, you'll focus on:Designing and deploying agent architectures that plan, act, and learn in complex environmentsImplementing continuous evaluation, safety, and learning loops for production-grade autonomyDefining the roadmap for agentic control systems and AI infrastructureEstablishing research and experimentation practices that drive measurable resultsAreas of ResponsibilityAgentic ArchitectureDesign multi-agent and hierarchical planners with tool use, memory, retrieval, and self-reflection (e.g., ReAct / Reflexion paradigms)Build tool-augmented agents for bidding, creative generation, analytics, and constraint solvingDevelop mechanisms for verification, recovery, and safe operation under uncertaintyLearning & Control SystemsApply contextual bandits and reinforcement learning for bidding and budget allocationImplement online learning with safety envelopes and reward shaping aligned to metrics like CPA, ROAS, and LTVBuild autonomy guardrails: policy constraints, anomaly detection, cost caps, and human-in-the-loop controlsEvaluation & SimulationCreate offline, shadow, and canary pipelines for iterative testing and deploymentDevelop simulators, evaluation harnesses, and red-team frameworks for robustness and drift detectionLLM SystemsIntegrate retrieval-augmented generation for keyword and creative workflowsApply program synthesis and prompt orchestration to coordinate complex tool plansOptimize for latency and cost using adapter, distillation, and caching strategiesScience to ProductCollaborate with product and design to frame objectives, define metrics, and run experimentsTranslate research breakthroughs into capabilities customers can toggle, audit, and trustKey OutcomesA production-grade agentic loop that outperforms baseline bidding and pacing modelsA hardened safety and evaluation harness with datasets, guardrails, and rollback protocolsAn LLM+tools pipeline that automates creative and keyword generation with traceable reasoningClear documentation, experiment logs, and a hiring plan for future AI/ML rolesWhat We're Looking ForMust-Haves5–7+ years of applied ML experience with shipped systems impacting real-world KPIsExpert Python and hands-on PyTorch experience across training, serving, and iterationProven experience in at least two of the following areas:Agentic LLM systems (tool use, planning, memory)Reinforcement learning or contextual banditsRetrieval and rankingTime-series or anomaly detectionCausal inference or experimentationStrong background in statistics, experiment design, and counterfactual reasoningDemonstrated ability to build safe, autonomous systems aligned with business goalsNice-to-HavesExperience with ads, marketplaces, or other constrained control systemsFamiliarity with distributed systems (Ray, Triton, Kafka, feature stores, vector DBs)Experience with FastAPI, async Python, Kubernetes, SQL, and modern data tools (DuckDB, Polars, Pandas)Publications or OSS contributions in agentic AI, RL, or NLP (impact > pedigree)What Makes You Stand OutYou love the intersection of autonomy and safety — you ship agents that act, verify, and recoverYou move seamlessly from research to production, choosing "simple that wins" until cutting-edge is simplestYou define rewards and objectives clearly — and defend them with dataYou're comfortable owning architecture while writing production code dailyYou're driven to build self-learning, self-correcting systems that scaleWhat's in It for YouA defining role shaping the intelligence core of an emerging AI companyDirect collaboration with the founders on AI, product, and strategyCompetitive compensation and meaningful equityFlexible, remote-friendly working with US/EU overlapA culture built on trust, experimentation, and technical excellenceWhy Work With MeritMerit connects top AI talent with high-growth, venture-backed companies — giving you direct access to founders and roles that define industries. Spend 20 minutes with Merit, and get matched with opportunities like this one that align with your experience and ambitions. Access exclusive, high-impact AI roles Fast-tracked introductions to founders and hiring managers Guidance from your personal AI recruiter, available 24/7Ready to define the future of agentic AI systems?Meet Merit today and get matched with roles that match your ambition and potential.Apply via Merit to learn more about this Founding Lead AI Scientist opportunity.