Technical Architect Machine Learning

Job description

**Role & Responsibilities:** - Design and architect multi-layered ML solutions using Azure AI Foundry and ensuring they are robust, scalable, and production-ready. - Defining and implementing robust integration strategies between legacy control software (e.g., VB6) and modern Azure ML platforms, ensuring seamless data flow, scalability, and strict data security governance. - Architecting, developing, and deploying advanced machine learning and deep learning models specifically for curve prediction and automated quality scoring. - Designing and implementing sophisticated optimization engines to automate precision movements with micron-level accuracy. - Leading the transition from manual operator heuristics and rule-based systems to automated, data-driven logic and decision-making processes. - Design and architect multi-layered Agentic AI solutions using Azure AI Foundry and LLM Orchestration frameworks (e.g., LangGraph, Semantic Kernel). - Define the technical roadmap for "Plan-and-Execute" agentic loops, specifically for unstructured data extraction and autonomous system updates. - Lead model selection and evaluation strategies, comparing high-reasoning models like Claude 3.5 Sonnet with open-weight alternatives (Llama 3.1) for cost and performance optimization. - Performing extensive data pre-processing on historical calibration data to train and validate models, and building robust production monitoring, alerting, and retraining scripts. - Continuously enhancing, modifying, optimizing, and maintaining existing ML models to improve performance, accuracy, and efficiency. - Working closely with external and internal stakeholders to define requirements for ML use cases, gather feedback, and drive solution enhancements. - Generating actionable insights from large datasets and effectively communicating complex analysis results and model performance to clients and internal stakeholders. - Provide technical leadership to MLEs and Platform engineers to ensure architectural alignment across workstreams. **Skills Expectation:** - **Expertise in Python and SQL with PySpark for large-scale data processing.** - Experience Expertise in Tree-based models (e.g., XGBoost, LightGBM). - **10+years Strong foundation in Traditional ML algorithms (e.g., Logistic Regression, Cluster Analysis, Decision Trees, Statistical Modeling, Predictive Analysis, Regression, Classification).** - Experience with Encoder-Decoder architectures for sequence modeling. - **Proficiency in Deep Learning Architectures specifically for time series and curve data.** - **Deep expertise in Agentic AI, Time Series forecasting, Descriptive Machine Learning, and Exploratory Data Analysis.** - **Strong experience with Google Cloud Platform (GCP) Or Azure Cloud Services for ML workloads.** - **Expertise in LLM Orchestration patterns (ReAct, Chain-of-Thought, multi-agent collaboration).** - **Deep experience with Azure AI services, including AI Search, AI Foundry, and Azure OpenAI.** - Proven ability to design, develop, and deploy production-grade Machine Learning and Generative AI systems that deliver measurable ROI and contribute to long-term AI roadmaps. - Experience with advanced optimization techniques for black-box functions, including Bayesian Optimization and sophisticated hyperparameter tuning methods. - Experience with Databricks for ML workflows and data engineering. - Must have experience in putting models into production. - Experience in post-production deployment, including MLOps practices (monitoring, retraining, versioning), would be a significant plus. - Experience working with scalable, highly-interactive, high-performance ML systems and projects. - Great analytical skills with meticulous attention to detail. - Strong stakeholder and team management skills. - Experience with AI Governance, compliance, and risk mitigation strategies. - Proven experience in deploying and monitoring LLMs in a production environment. - Familiarity with containerization (Docker, Kubernetes) for ML model serving.