Design AI-native experiences for interacting with data via natural language, agents, assistant-style interfaces, and programmatic/CLI access.
Build prototypes and production workflows for KPI lookup, report-health checks, quality validation, insight routing, and stakeholder self-service.
Partner with Data Engineers to ground AI experiences in trusted semantic models and governed data assets.
Work with Business Analysts to encode business context, metric interpretation, ontology, and guardrails into AI experiences.
Support AI-native skilling and engineering practices, including human-in-the-loop needs and capability gaps.
Drive automation opportunities such as reporting automation, alert/response, validation, and insight routing.
Apply responsible AI, security, entitlement, and governance principles throughout.
Define reusable architecture patterns for AI-enabled decision intelligence.
Yêu cầu công việc
Must-Have Qualifications
Core requirements:
Excellent English communication — able to clearly understand spoken and written English, and express ideas and collaborate in confident, fluent, and precise English.
Growth mindset and drive — eager to learn and improve, proactive and self-motivated, self-critical and open to feedback, while inclusive and self-assured.
Alignment and reliable execution — stays aligned with company and team strategy and roadmap, and dependably executes and delivers on commitments.
Role-specific:
A solid software engineering foundation — system design, data structures, concurrency, error handling, and building services that are reliable in production. You treat AI systems with the same engineering rigor as any other software you ship.
Strong programming skills, ideally in Python and/or TypeScript/C#. You're comfortable working close to the metal — for example, implementing a tool-calling loop yourself rather than depending on a framework to handle it for you.
Practical experience building AI applications, agents, assistants, or LLM-integrated workflows, and the ability to clearly explain what you built, the decisions you made, and what you'd do differently next time.
A real grasp of the fundamentals beneath the frameworks: how APIs, backend services, and orchestration fit together; how prompts and context are structured and managed within a model's limits; and how tool/function calling works.
The ability to debug and evaluate AI systems — to figure out why a model's output is wrong, and to measure whether a change genuinely improved things rather than relying on gut feel.
Awareness of cost, latency, and reliability trade-offs, and the judgment to choose the simplest approach that solves the problem well.
Good instincts for matching the technique to the problem — retrieval, fine-tuning, prompt engineering, or plain function calling — based on what the work actually needs.
Experience-to-Have
Experience with the Microsoft technology stack (e.g. .NET/C#, Azure, Azure OpenAI, Microsoft Fabric).
Data engineering skills are a big plus — dimensional modeling (star and snowflake schemas), ETL pipelines, and working with big data.
Familiarity with enterprise data platforms and semantic layers.
Experience designing AI systems that are grounded, explainable, permission-aware, and aligned with governed data sources.
Standard engineering practice: Git, CI/CD, testing, logging, monitoring, and deployment.
