Job Summary

Data model design — the most important skill on this list. Designing the data structure that every report sits on top of: star schema / dimensional semantic modelling, built properly from messy operational sources rather than worked around inside the reports. 2. Advanced DAX — writing the calculations behind our compliance percentages and month-over-month trends, and getting them right: filter context, CALCULATE modifiers, time intelligence over a custom date table, calculation groups. 3. SQL Server — some of our data sits in a SQL database. You will write and tune queries against it, not just read from it: T-SQL, reporting views and stored procedures, execution plans, indexing. 4. Snowflake — our historical and archived data sits in a Snowflake warehouse. You will need to serve it in reports without running up compute cost. 5. API data extraction — one of our systems (Flexera One) releases data only through a REST API. You will build a scheduled extract that runs unattended: token-based authentication including renewal, service accounts, pagination, rate limits, retries, and secure credential storage such as Azure Key Vault. Output must land in a durable staging layer — SQL tables or a Fabric lakehouse — that the model then reads from. A model calling the API live at refresh time is not acceptable. 6. Combining the sources — joining all of the above into one trustworthy dataset, with the heavy lifting pushed back to the source systems: Power Query / M, query folding against both SQL and Snowflake, incremental refresh. 7. Performance engineering — our current reports are slow. You will need to measure why and prove the improvement with actual numbers, using Performance Analyzer, DAX Studio, VertiPaq Analyzer and Tabular Editor\'s Best Practice Analyzer. 8. Row-level security — each application owner or user must see information relevant to them and their role or access. Dynamic RLS driven by the signed-in user, tested with impersonation. 9. Copilot / AI readiness — users should be able to ask a question in plain English and get a correct answer. This is preparation work on the semantic model, not a feature toggle: AI instructions (\"Prep data for AI\"), verified answers, business-friendly naming and descriptions, and testing AI answers against known-correct results before calling it done. Production experience required. 10. Maintainable Documented handover — delivered as a versioned PBIP/PBIR project in Git with real documentation: model definitions, data lineage, and a refresh and troubleshooting runbook. Not a single PBIX file and a verbal walkthrough. on Power BI development, including, building models that combine multiple source systems \\\r\\\\n1. Data model design — the most important skill on this list. Designing the data structure that every report sits on top of: star schema / dimensional semantic modelling, built properly from messy operational sources rather than worked around inside the reports.\\\\r\\\\n2. Advanced DAX — writing the calculations behind our compliance percentages and month-over-month trends, and getting them right: filter context, CALCULATE modifiers, time intelligence over a custom date table, calculation groups.\\\\r\\\\n3. SQL Server — some of our data sits in a SQL database. You will write and tune queries against it, not just read from it: T-SQL, reporting views and stored procedures, execution plans, indexing.\\\\r\\\\n4. Snowflake — our historical and archived data sits in a Snowflake warehouse. You will need to serve it in reports without running up compute cost.\\\\r\\\\n5. API data extraction — one of our systems (Flexera One) releases data only through a REST API. You will build a scheduled extract that runs unattended: token-based authentication including renewal, service accounts, pagination

Key Responsibilities

NA

Skill Requirements

Power BI Data Model Design, Advanced DAX, SQL Server Snow Flake and API data extraction

Other Requirements

We have a dashboard that tracks whether our IT applications are running supported, up-to-date infrastructure (Hardware & software). It works, but one of the key data sources is changing to a SaaS solution and the current dashboard is slow, has grown too large, and was built before our current data platform and before AI features existed in Power BI. We need one senior person to rebuild it: pull data from four different systems, design a clean and fast data model underneath it, build a small number of genuinely useful reports, and set it up so users can ask questions in plain English and get correct answers. Our data lives in: a SQL Server database (current data) · a Snowflake warehouse (historical/archived data) · Flexera One (accessible only through an API) · ServiceNow (read-only) · SharePoint