AI Engineer - Data42

Job Description Summary

Data42 is Novartis’s data and analytics platform, integrating different data modalities at scale, including discovery, preclinical, clinical, and real-world data, to support data-driven decision-making. We are seeking an AI Engineer who can translate business and scientific problems into secure, scalable, and observable AI-enabled products. The role combines hands-on software engineering, machine learning engineering, agentic system design, and production operations. Experience with Palantir Foundry is advantageous but not mandatory, equivalent experience on modern cloud, data, and AI platforms is welcomed

Job Description

Key Responsibilities

• Design and deliver end-to-end AI-powered applications, from discovery and architecture through implementation, deployment, monitoring, and continuous improvement.

• Build agentic AI solutions using large language models, tool use, structured workflows, retrieval-augmented generation, memory and context patterns, orchestration, and human-in-the-loop controls.

• Engineer and productionize machine learning solutions, including data preparation, feature engineering, model training or adaptation, evaluation, deployment, inference, monitoring, and retraining workflows.

• Develop reliable backend services and APIs using Python and modern service patterns; build intuitive web experiences using TypeScript, React, and contemporary UI frameworks where required.

• Integrate enterprise data sources, APIs, vector stores, model endpoints, and business systems into secure AI applications and automated workflows.

• Design robust system architectures spanning frontend applications, backend services, data pipelines, ML services, agent orchestration, observability, and deployment infrastructure.

• Create evaluation frameworks for AI quality, grounding, task completion, safety, latency, cost, and business impact; use results to improve prompts, models, retrieval, tools, and workflows.

• Implement responsible AI and security controls, including access control, data protection, auditability, guardrails, traceability, human oversight, and compliance-by-design.

• Apply software engineering discipline through automated testing, version control, code review, CI/CD, infrastructure configuration, release management, and production support.

• Operate production AI services with appropriate logging, tracing, alerting, incident response, performance optimization, reliability controls, and service documentation.

• Partner with product managers, data scientists, domain experts, architects, UX designers, platform teams, and governance stakeholders to convert complex requirements into measurable product outcomes.

• Provide technical guidance, reusable patterns, documentation, and engineering best practices that accelerate safe adoption of AI across teams.

Minimum Qualifications

• Bachelor’s degree in Computer Science, Engineering, Data Science, or a related technical field, a Master’s degree is preferred.

• Typically 5+ years of professional software, data, or ML engineering experience, including meaningful hands-on delivery of AI/ML applications in production environments.

• Strong proficiency in Python and practical experience with software architecture, APIs, data structures, testing, and maintainable application design.

• Hands-on experience implementing machine learning lifecycle activities such as experimentation, feature/data pipelines, model evaluation, deployment, inference, and monitoring.

• Hands-on experience building applications with LLMs or agentic AI patterns, including prompt and context design, tool integration, retrieval, workflow orchestration, evaluation, and guardrails.

• Experience with cloud-native or enterprise platforms and production engineering practices, including containers, CI/CD, environment management, secrets, identity and access controls, and observability.

• Working knowledge of modern data technologies such as SQL, Spark or distributed processing, data lakes or warehouses, and heterogeneous data integration.

• Ability to design for non-functional requirements including security, privacy, scalability, reliability, maintainability, performance, and cost.

• Proven ability to translate ambiguous business or scientific needs into technical designs, working increments, measurable acceptance criteria, and production outcomes.

• Clear communication, collaboration, and documentation skills, with a product-minded and …