Overview

Senior Machine Learning Engineer, AI Studio

Responsibilities

• Define the user, workflow, decision, intended use, baseline, value hypothesis, acceptance criteria, adoption path, operating owner and measurable technical and business outcomes.
• Map rules, exceptions, data dependencies and human decision points before selecting deterministic automation, classical ML, deep learning, GenAI, RAG, agents or a manual approach.
• Own production architecture across data, feature and knowledge pipelines, models, retrieval, agents, APIs, persistence, workflows, user experience, security zones and human review.
• Lead hands-on development of production software, EDA, feature engineering, predictive models, deep-learning or NLP components, inference services, RAG, agent tools and workflow orchestration.
• Establish baselines, experiment design, leakage controls, uncertainty, calibration, subgroup and robustness checks, gold sets, error taxonomies, expert adjudication and release thresholds.
• Establish MLOps/LLMOps for lineage, reproducibility, versioning, CI/CD, canary or shadow release, observability, drift monitoring, SLOs, rollback, incidents, disaster recovery, capacity, cost and runbooks.
• Coordinate security, privacy, Responsible AI, Quality, legal, model-risk and GxP controls; create reusable capabilities, measure adoption and value, mentor engineers and improve delivery practices.

Qualifications

• Bachelor’s/Master’s degree with 8 - 13 years of experience in Computer Science, IT or related field.
• Advanced software and AI/ML system design: Production Python and SQL, APIs, distributed or event-driven services, data persistence, testing, performance, repository governance, design review and end-to-end architecture.
• Advanced statistics, ML and experimental design: EDA, feature engineering, supervised and unsupervised learning, predictive modelling, ensembles, anomaly detection, calibration, uncertainty, robustness, explainability and causal reasoning where justified.
• Deep learning, NLP and foundation models: Selection and production use of neural, transformer, embedding, vision, document and multimodal approaches, including fine-tuning versus prompting, latency, privacy and cost trade-offs.
• GenAI, RAG, knowledge and agentic AI: Grounded retrieval, structured output, provenance, citations, abstention, entitlement controls, bounded tool use, permissions, durable state, recovery, adversarial evaluation and human control.
• Data, knowledge, cloud and AI operations: Trustworthy batch/streaming pipelines, data contracts, lineage, vector/graph stores, Spark or Databricks, containers, Kubernetes, MLOps/LLMOps, SLOs and lifecycle operations.
• Responsible AI and regulated delivery: Risk assessment, least privilege, threat modelling, red teaming, bias and subgroup robustness, privacy, model/data documentation, human accountability, validation and applicable GxP controls.
• Demonstrated end-to-end ownership of at least one production ML, GenAI, software, data or automation system that delivered a measurable outcome.
• Strong hands-on proficiency in Python and SQL, with experience designing production software, services and evaluation pipelines.
• Advance capability in at least one role-defining pillar—Applied ML, GenAI/RAG/agents or ML platform/MLOps—plus credible depth across the production lifecycle.
• Advanced ML, causal and uncertainty methods: Experience with data-centric AI, weak supervision, active learning, conformal or Bayesian uncertainty, causal inference, time-series, survival methods or drift-aware retraining.
• Advanced deep learning and model efficiency: Experience with transformers, multimodal pipelines, CNNs, RNNs, GNNs, PEFT or LoRA, fine-tuning, distillation, quantization, routing, cascades or inference optimization.
• Cloud, platform and AI operations: Experience with AWS, Bedrock or SageMaker, Databricks, Spark, Kubernetes, serverless systems, infrastructure as code, MLflow, Airflow, Kubeflow, observability and FinOps.
• Human-AI and regulated delivery: Experience with review, correction, approval, accessibility, uncertainty communication, workflow automation and GxP-relevant or validated systems.
• Strong product thinking and ability to connect technical decisions to user, workflow, risk, cost and business value.
• Technical leadership, mentoring and constructive challenge while remaining hands-on.
• Excellent analytical judgment and clear communication of evidence, uncertainty, trade-offs and limitations.
• Cross-functional leadership across business, product, architecture, engineering and control functions.
• Ownership, resilience and continuous improvement through incidents, feedback and measured outcomes.