DP-800 Practice Questions & Exam Prep — Microsoft SQL AI Developer
Get exam-ready for DP-800 (Microsoft Certified: SQL AI Developer Associate) on iPhone or iPad. Azure Mastery uses on-device AI to predict your readiness score across all three DP-800 domains — Design & develop, Secure & optimize & deploy, and AI capabilities in SQL — builds a personalised study plan from your weak spots, and surfaces topics you're forgetting. Core study stays on-device and works offline; optional sync uses your private iCloud account.
The exam
What is the DP-800 exam?
DP-800 is the exam that earns the Microsoft Certified: SQL AI Developer Associate credential (exam title: "Developing AI-Enabled Database Solutions") — the credential Microsoft introduced for the role hiring managers now post as "SQL AI Developer", "Database Developer (AI)", or "Fabric Data & AI Developer". It's the natural next step after DP-900 for SQL developers whose day job has expanded from classical schema work into shipping embeddings, vector search, and retrieval-augmented generation (RAG) directly inside SQL Server, Azure SQL, and SQL databases in Microsoft Fabric. DP-800 is the only Microsoft Associate cert dedicated to AI-enabled data-tier development.
DP-800 is code-first and developer-focused. It validates that you can design tables and indexes (including column store, in-memory, temporal, ledger, external, and graph) and choose the right constraints, sequences, and partitioning; write advanced T-SQL — CTEs, window functions, JSON shaping (JSON_OBJECT/OPENJSON/JSON_MODIFY), regex (REGEXP_LIKE/REPLACE/SUBSTR/INSTR), fuzzy matching (EDIT_DISTANCE, JARO_WINKLER_DISTANCE), graph MATCH queries, error handling; secure data with Always Encrypted, Dynamic Data Masking, Row-Level Security, Managed Identity passwordless access; tune workloads via Query Store and DMVs; deploy with SQL Database Projects and Data API Builder; and bring AI into SQL itself — generating embeddings, building vector indexes, designing RAG flows with sp_invoke_external_rest_endpoint, structuring outputs with JSON. Expect scenario questions that show a snippet of T-SQL or a chunking strategy and ask what's wrong, what's better, or what's required.
Microsoft published the DP-800 skills outline on 12 March 2026. Every question in Azure Mastery's DP-800 bank is mapped to the current outline — no questions on retired services or pre-AI SQL-only patterns. Read the official outline at learn.microsoft.com.
Questions40–60 mixed format
Duration100 minutes (120 min seat)
Pass score700 / 1000
CostUSD $165 (≈ £128 UK)
ValidityRenew annually (Associate)
FormatOnline or test centre
Skills measured · March 2026
DP-800 exam objectives
Three domains, with weights set by Microsoft's 12 March 2026 outline. Every domain summary below is paraphrased from the official skills outline; bullet-level objectives in Azure Mastery are tagged so you always know which domain you're being tested on and where your weak spots cluster.
Data skill mapSourcesTransformStoreAnalyse
Design and develop database solutions35–40%
Explore key topics
The schema and code foundation. Tables and indexes — data types, sizing, column store, in-memory (memory-optimized), temporal, ledger, external (PolyBase / Azure SQL external), and graph (node/edge). JSON columns and indexes, computed JSON columns. Constraints (PRIMARY KEY, FOREIGN KEY, UNIQUE, CHECK, DEFAULT), SEQUENCES, partitioning. Programmability — views (regular and indexed), scalar and table-valued functions, stored procedures, triggers (INSTEAD OF vs AFTER, DML vs DDL). Advanced T-SQL — CTEs (recursive and non-recursive), window functions, JSON_OBJECT/OPENJSON/JSON_MODIFY, regex (REGEXP_LIKE/REPLACE/SUBSTR/INSTR), fuzzy matching (EDIT_DISTANCE, JARO_WINKLER_DISTANCE), graph MATCH queries, correlated queries, TRY/CATCH error handling. AI-assisted development — GitHub Copilot, Microsoft Copilot in Fabric, Model Context Protocol (MCP) tool configuration, Copilot instruction files, and the security impact of each. Around 14–20 questions per sitting.
Secure, optimize, and deploy database solutions35–40%
Explore key topics
The production-readiness surface. Security — Always Encrypted with secure enclaves, column-level encryption, Dynamic Data Masking, Row-Level Security (RLS), object-level permissions, auditing, passwordless access via Managed Identity, key management with Azure Key Vault. Performance — DMVs, Query Store, Query Performance Insight, execution-plan analysis, isolation-level trade-offs for the workload. Deployment — SQL Database Projects for CI/CD on GitHub, schema-drift detection, secrets management, branching policies, reference/static data in source control. Integration — Data API Builder (DAB) REST/GraphQL endpoints, Application Insights, Log Analytics, change tracking, change data capture (CDC), change event streaming (CES), Azure Functions / Logic Apps. Around 14–20 questions.
Implement AI capabilities in database solutions25–30%
Explore key topics
The AI-in-the-data-tier story. Register external AI models from SQL. Generate and maintain embeddings; design chunking strategies that preserve semantic boundaries. Build vector indexes for nearest-neighbour search, choose between full-text, semantic-vector, and hybrid search (including reciprocal rank fusion) for the workload. Build retrieval-augmented generation (RAG) flows using sp_invoke_external_rest_endpoint — JSON shaping for language-model input, structured-output parsing on the way back. Decide where the AI primitive belongs (SQL vs Azure AI Search vs application layer) given latency, governance, and cost. Around 10–18 questions.
Designed for DP-800
How Azure Mastery helps you pass DP-800
Azure Mastery ships with 364 DP-800 practice questions, every one written specifically against the current (March 2026) skills outline. Each question carries a domain tag mapped to the official three domains (Design & develop, Secure & optimize & deploy, AI capabilities in SQL), so you always know which area you're being tested on and where your weak spots are clustered. T-SQL snippets — CTEs, window functions, JSON shaping, regex, fuzzy matching, graph MATCH queries, vector index DDL, RAG flow stubs — appear throughout, matching the format of the live exam.
The on-device Exam IQ engine predicts your DP-800 score before you sit the exam. After roughly 30 questions it has enough signal to give a confidence-scored prediction (e.g. "786 ±37, 68% confidence") — and tells you the specific topics that are dragging your readiness down. No vague "study more" advice; just a ranked list of objectives where improvement would move your score the furthest.
The adaptive study plan rebuilds itself from your answer history. Get an Always Encrypted vs Dynamic Data Masking scenario wrong? You'll see another data-protection-choice question in the next session. Master "full-text vs vector vs hybrid search with reciprocal rank fusion" three sessions running and the engine backs off, surfacing fresh JSON_MODIFY or SQL Database Projects scenarios. The plan optimises for the gap between where you are and the 700 pass score, not for blind volume.
Knowledge decay tracking matters for DP-800 because the AI domain (25–30%) moves quickly — vector index types, embedding model trade-offs, and RAG patterns evolve faster than classical T-SQL knowledge. Azure Mastery tracks every topic's decay curve and flags topics approaching expiry. The padlock icon on the Today screen is your "revisit before you forget" cue, and weak-spot drills automatically pull from decayed topics first.
Real exam simulation mode runs at DP-800's actual length and time pressure: a randomised 40–60-question set drawn from the full 364-question bank, weighted by domain percentages from the March 2026 outline, with the 100-minute timer running and no jumping back to flag-and-review. It's the closest you can get to the live Pearson VUE / online-proctored experience without sitting the exam.
Answer Coach turns each missed answer into a private, grounded lesson: the misconception, key distinction, and rule to remember. It always uses authored certification guidance; on supported devices, an optional on-device model may rewrite the note only when it passes grounding checks.
During your first week, Aura adapts the next step as you go. Every session ends with a concise recap of what changed, what to focus on, and the best follow-up.
Everything essential runs on-device. Your answer history, readiness gauge, and coaching stay private. Optional sync uses your private iCloud account; there is no Azure Mastery account, tracking, or external processing server.
6-week study plan
Suggested DP-800 study plan
Most candidates pass DP-800 after six to ten weeks of focused study, depending on prior T-SQL fluency and exposure to embeddings / vector search. The six-week plan below maps onto the three DP-800 domains (the first two weight 35–40% each), Azure Mastery's adaptive sessions, and the in-app exam simulator. Adjust pace to taste — the readiness gauge tells you when you're done, not the calendar.
Design and develop database solutions
Days 1–3: Tables and indexes — data types, sizing, column store, in-memory (memory-optimized), temporal, ledger, external (PolyBase / Azure SQL external), and graph (node/edge). JSON columns and indexes, ISJSON, JSON_PATH_EXISTS, computed JSON columns.
Days 4–6: Constraints (PRIMARY KEY, FOREIGN KEY, UNIQUE, CHECK, DEFAULT), SEQUENCES, partitioning (partition functions, schemes, sliding window). Programmability — views (regular and indexed), scalar / table-valued functions, stored procedures (parameter sniffing, recompilation), triggers (INSTEAD OF vs AFTER).
Days 7–10: Advanced T-SQL — CTEs (recursive and non-recursive), window functions (ROW_NUMBER, RANK, LAG/LEAD, SUM OVER, frame specifications), JSON_OBJECT/JSON_ARRAY/OPENJSON/JSON_MODIFY shaping.
Days 11–14: Regex functions (REGEXP_LIKE/REPLACE/SUBSTR/INSTR/COUNT/MATCHES), fuzzy matching (EDIT_DISTANCE, JARO_WINKLER_DISTANCE), graph MATCH queries, correlated queries, TRY/CATCH/THROW/RAISERROR. AI-assisted development — GitHub Copilot, Microsoft Copilot in Fabric, MCP tool configuration, Copilot instruction files, and the security impact of each.
Secure, optimize, and deploy database solutions
Days 15–17: Security at rest and in motion — Always Encrypted (column-level, with and without secure enclaves), Dynamic Data Masking, Row-Level Security (RLS), object-level permissions, auditing.
Days 18–20: Passwordless access via Managed Identity, key management with Azure Key Vault, authentication patterns for SQL Server / Azure SQL / SQL in Fabric.
Days 21–23: Performance — DMVs, Query Store, Query Performance Insight, execution-plan analysis, isolation-level choice for the workload. Diagnose plan regressions and blocking.
Days 24–28: Deployment with SQL Database Projects for CI/CD on GitHub — branching, schema drift, secrets management, reference/static data in source control. Integration — Data API Builder (DAB) REST/GraphQL endpoints, Application Insights, Log Analytics, change tracking, CDC, change event streaming (CES), Azure Functions and Logic Apps.
AI capabilities in SQL, sharpen, simulate
Days 29–32: Register external AI models from SQL. Generate and maintain embeddings. Design chunking strategies that preserve semantic boundaries (and don't.) Build vector indexes for nearest-neighbour search.
Days 33–36: Choose between full-text, semantic-vector, and hybrid search — including reciprocal rank fusion. Build RAG flows with sp_invoke_external_rest_endpoint, JSON shaping for language-model input, structured-output parsing on the return.
Days 37–40: Run Focus Weak Spots every morning. The two heaviest domains (Design & develop, Secure & optimize & deploy) weight 35–40% each — give them roughly equal time; do not let the smaller AI domain crowd them out.
Days 41–42: Two end-to-end Exam Simulator runs at full 100-minute length. Review carefully. If readiness gauge is 750+ with reasonable confidence, schedule the exam.
Inside the app
Every Microsoft question type, on iPhone
DP-800's question bank uses the same formats Microsoft puts on the live exam — not just multiple choice. Each visualisation below is a faithful mock of how the type renders inside Azure Mastery on iPhone and iPad. Exam-simulator mode runs all of them at full 100-minute length with no flag-and-review jumps, mirroring Pearson VUE.
Which HTTP method does a RAG procedure use to invoke a chat-completions endpoint?
POST
GET
DELETE
PATCH
Multiple choice
A real DP-800 question-bank example with one correct answer. The app explains every option after you answer.
Exam-specific sample
Which two features help you retain and review historical changes to rows over time? (Select two.)
System-versioned temporal tables
Change data capture (CDC)
A nonclustered index on the primary key
A computed column that uppercases a name
A check constraint on a numeric range
Multi-select
A real DP-800 multi-select item. Every required selection must be correct to earn the mark.
All-or-nothing
Order the steps to build an in-database retrieval-augmented generation pipeline in Azure SQL.
⋮⋮1Register the embedding model with CREATE EXTERNAL MODEL
⋮⋮2Chunk the source text and generate embeddings for each chunk
⋮⋮3Create a vector index on the embedding column
⋮⋮4Run VECTOR_SEARCH to retrieve the most similar chunks
⋮⋮5Pass the retrieved chunks to a chat model to compose the answer
Drag-and-drop
A real DP-800 interactive-format prompt, rendered for touch on iPhone and iPad.
Interactive item
On the Azure SQL logical server's Microsoft Entra settings, which option should you enable so that SQL authentication logins can no longer connect and only Entra…
Hotspot
A real DP-800 prompt that tests recognition inside a visual or contextual interface.
Tap target
Contoso Health — Securing a regulated SQL AI application Contoso Health runs a patient-records application on Azure SQL Database and is adding AI features that summarize visit notes. Regulators require that sensitive…
1Which feature keeps the national-identifier column unreadable even to database…
2How can the database ensure each clinician queries only their own patients' rows?
3Which feature shows support staff a masked identifier without altering the…
4How should the App Service tier connect to Azure SQL without a stored database…
Case studies
A real DP-800 case-study scenario with linked questions that share the same requirements and environment.
Multi-question
✕Your answer: GET
✨ Answer Coach:GET retrieves resources and carries no request body, so it cannot send the JSON payload a completions call requires.
— grounded in authored certification guidance
Answer Coach
Answer Coach uses the bank's authored rationale to explain the misconception, key distinction, and rule to remember. On supported devices, an optional on-device model may rewrite the note only when it passes grounding checks.
App exclusive
Frequently asked
DP-800 FAQs
How much does the DP-800 exam cost?
The DP-800 voucher is USD $165 in the United States. Pricing varies by region — in the UK it's typically around £128. Microsoft sometimes runs free-voucher promotions during events such as Microsoft Build or Microsoft Ignite, so check your Microsoft Learn profile for any active offers before booking. DP-800 is an Associate-tier credential and requires annual renewal (free, online), so factor that into long-term cost planning.
Does the DP-800 certification expire?
Yes. Microsoft Associate certifications including DP-800 expire annually. Renewal is free — a short online assessment on Microsoft Learn within the six-month window before your expiration date. The renewal targets recent skills outline updates, so staying current is straightforward if you remain broadly active building on SQL Server, Azure SQL, and SQL in Microsoft Fabric. (Fundamentals certifications such as DP-900 are different — those don't expire.)
What is the DP-800 retake policy if I fail?
The first retake is allowed after 24 hours. Second and third retakes each require a 14-day wait. Microsoft caps retakes at five attempts per 12-month rolling period. Each attempt requires a new voucher purchase. DP-800 retake failures usually come from underestimating the AI domain — most candidates over-study T-SQL and JSON shaping and under-study embeddings, vector indexes, and RAG flow design.
How long should I study for DP-800?
Most candidates pass DP-800 after six to ten weeks of focused study, assuming working T-SQL fluency and basic awareness of embeddings and vector search. If T-SQL is new to you, plan for three months — the exam expects you to write CTEs, window functions, JSON shaping queries, and regex/fuzzy-match patterns from memory, not just describe what they do. Azure Mastery's readiness gauge tells you when you're at exam-ready; don't book until it shows roughly 750 or higher with reasonable confidence.
DP-800 vs DP-300 — which should I take?
Different jobs. DP-300 (Microsoft Azure Database Administrator Associate) is for DBAs — running, securing, tuning, and recovering Azure SQL workloads (Azure SQL Database, Managed Instance, SQL on VM) including HA/DR, performance tuning, and automation. DP-800 is for SQL developers bringing AI primitives — embeddings, vector search, RAG, MCP endpoints — directly into the data tier through T-SQL. DP-300 administers; DP-800 develops. Many SQL pros take both because AI-enabled apps need both DBA discipline and developer fluency.
DP-800 vs AI-103 — what's the difference?
Different stacks. AI-103 is for building generative AI solutions and agents with Microsoft Foundry, including model selection, orchestration, tools, knowledge sources, and operational readiness. DP-800 is for SQL developers bringing AI primitives such as embeddings, vector indexes, RAG, and structured outputs directly into SQL Server, Azure SQL, and Microsoft Fabric with T-SQL. AI-103 is application and agent-layer; DP-800 is data-layer.
Where DP-800 fits
Certification paths that include DP-800
DP-800 is the SQL AI Developer Associate cert — the developer counterpart to DP-300's DBA track. It pairs with DP-900 as the recommended fundamentals, and bridges into application and agent-layer AI work via AI-103. Tap any linked exam below to see its dedicated study app page.
DP-800 sits at the data-platform Associate tier as the developer counterpart to DP-300. DP-900 builds the data fundamentals it leans on; DP-300 (DBA) and AI-103 (generative AI apps and agents) provide adjacent role coverage.
Ready to pass DP-800?
Download Azure Mastery free. 364 DP-800 practice questions across all three domains, AI score prediction, full-length exam simulator, adaptive study plan. iPhone & iPad.
