Chief AI/ML Architect – Causal AI & Decision Systems

ppbLocation: /b Zurich, Switzerland / Hybrid / Remote /ppbRole Type: /b Full-Time /ppbDomain: /b Enterprise AI Operating Systems / Causal Machine Learning / Agentic AI / Complex Physical Systemsbr/ /ph3About the Role /h3pWe are seeking a bChief AI/ML Architect /b to design and lead the end-to-end technical vision of our next-generation bCausal Decision Operating System /b. /ppMost of the enterprise AI world remains trapped in passive statistical prediction—learning surface-level correlations that break down when market conditions change—or deploying unstructured text wrappers that lack operational discipline. We are building a platform that unifies bCausal Machine Learning /b (understanding root causes, interventions, and "what-if" counterfactuals) with bAutonomous AI Reasoning /b (agentic workflows, neuro-symbolic planning, and natural language interfaces). /ppIn this executive engineering role, you will architect the entire AI intelligence layer. You will bridge low-level mathematical causal inference with high-level agentic orchestration, allowing our platform to translate plain-English enterprise queries into structural causal models, evaluate interventions against hard financial guardrails, and execute closed-loop actions across corporate infrastructure.br/ /ph3Key Responsibilities1. Unified AI Architecture Reasoning Systems /h3ullibCausal Engine Architecture: /b Lead the design of scalable systems for automated Causal Graph Discovery, Structural Causal Models, Treatment Effect Estimation, and high-throughput Counterfactual Simulation engines. /lilibAgentic Workflows Symbolic Planning: /b Architect autonomous AI agents and neuro-symbolic reasoning frameworks capable of executing multi-step operational plans while remaining strictly bound to deterministic enterprise guardrails (such as double-entry accounting rules, inventory conservation, and physical plant safety limits). /lilibNatural Language to Knowledge/Causal Graphs: /b Design systems that allow business executives to ask complex operational queries in natural language, translating those requests into executable structural causal models and returning audited, explainable recommendations. /li /ulh32. Robustness Autonomous Execution /h3ullibDomain Generalization Distribution Shifts: /b Implement Invariant Causal Prediction and out-of-distribution generalization frameworks so the AI system adapts reliably during supply chain shocks, macro economic shifts, or physical sensor failures. /lilibClosed-Loop Action Orchestration: /b Build safe execution pipelines that move the platform from passive decision-support to automated action dispatching across core enterprise systems (ERP, WMS, TMS, MES). /li /ulh33. Enterprise Infrastructure MLOps/AIOps /h3ullibProduction Systems Strategy: /b Oversee the architectural blueprint for streaming data ingestion, online/offline graph updates, automated model refutation, and zero-downtime agent deployment inside secure client cloud environments (VPC / On-Premise). /lilibHigh-Dimensional Data Pipelines: /b Architect pipelines that merge high-frequency physical time-series (sensors, logs) with transactional enterprise databases (such as SAP ledgers, Snowflake, or custom data lakes). /li /ulh34. Technical Leadership Strategy /h3ullibDefine Technical Vision: /b Act as the final authority on system architecture, machine learning methodology, model safety, and software engineering standards across the organization. /lilibTranslate Research into Enterprise Software: /b Benchmark and adapt cutting-edge research in causal inference, graph neural networks, and agentic reasoning from premier research institutes into production-ready platform capabilities.br/ /li /ulh3Required Qualifications Expertise /h3h31. AI Reasoning, Agentic Workflows Neuro-Symbolic AI /h3ulliDemonstrated track record in designing bAutonomous Agentic Architectures /b and multi-step planning systems for enterprise software. /liliPractical mastery of bNeuro-Symbolic Integration /b: combining probabilistic AI/deep learning outputs with deterministic, rule-based symbolic logic. /liliExpertise in bGraph Representation Learning /b (Graph Neural Networks, Relational GCNs) and enterprise Knowledge Graphs. /liliExperience building natural language interfaces that query complex structured data or probabilistic graphs without hallucination risks. /li /ulh32. Causal Machine Learning (Core Intelligence) /h3ulliDeep theoretical and practical command of bStructural Causal Models /b and Pearl’s Ladder of Causation. /liliProven experience with bCausal Discovery Algorithms /b applied to observational time-series and high-dimensional transactional data. /liliAdvanced knowledge of bTreatment Effect Estimation /b (Heterogeneous Treatment Effects, Individual Treatment Effects, Double Machine Learning, Causal Forests) using specialized frameworks. /liliFamiliarity with bCausal Refutation Protocols /b (such as dummy outcome tests, placebo treatments, and unobserved confounder sensitivity testing). /li /ulh33. Production Systems Engineering Infrastructure /h3ullib8+ years /b of software engineering experience with at least b4+ years /b architecting distributed, production-grade AI/ML platforms. /lilibTech Stack Mastery: /b Python (PyTorch, Ray, PyTorch Geometric, Polars/DuckDB), C++, Distributed Systems (Kubernetes, Spark/Flink), Streaming (Kafka/Redpanda), and Cloud Infrastructure (AWS/Azure/GCP, Terraform, Docker). /liliExperience deploying containerized, security-hardened AI architectures directly into enterprise client networks.br/ /li /ulh3Preferred Background /h3ullibAcademic Credentials: /b Ph.D. or Master’s degree in Computer Science, Artificial Intelligence, Machine Learning, or Applied Mathematics from a top-tier research institution (e.g., ETH Zurich, EPFL, Max Planck Institute, Cambridge, Oxford, MIT, Stanford). /lilibEnterprise Tech Background: /b Prior leadership experience at scaleups or enterprise software vendors specializing in process mining, supply chain intelligence, platform orchestration, or digital twin architecture. /li /ul /p #J-18808-Ljbffr