Certifications

The certifications that matter for AI engineering and data engineering, by vendor and level, with what each proves and where to learn for it here. Where we have free prep (study guides, practice and mock exams), it is marked.

AI Engineering Data Engineering By vendor Free prep: Claude Free prep: Azure AI-103 Free prep: Databricks GenAI Free prep: Databricks Data Engineer Professional

AI Engineering

AI engineering certifications

For people who build products and agents with models, and for those who train and run the models.

VendorCertificationLevelWhat it provesLearn for it here
AnthropicClaude Certified Architect – Foundations (CCA-F)
free prep here
FoundationalDesigning with Claude: agents, Claude Code, tools and MCP, prompts, contextAgent Harness course
AnthropicClaude Certified Architect – Professional (CCAR-P)
free prep here: guide line by line
ProfessionalThe next level of the same: production agent architectureAgent Harness Series
MicrosoftAI-901: Azure AI FundamentalsFundamentalsAI ideas and the Azure AI services (replaced AI-900 in 2026)The AI Universe
MicrosoftAI-103: Developing AI Apps and Agents on Azure
free prep here
AssociateBuilding AI apps and agents on Azure (replaced AI-102 in 2026)AI path on Azure
MicrosoftAI-300: MLOps engineeringAssociateTraining, deploying and running models on Azure (replaced DP-100 in 2026)ML path on Azure
AWSAWS Certified AI Practitioner (AIF-C01)FoundationalAI, ML and generative AI ideas and the AWS services for themThe AI Universe
AWSMachine Learning Engineer – Associate (MLA-C01)AssociateBuilding, deploying and monitoring ML on AWSML path on AWS
AWSGenerative AI Developer – Professional (AIP-C01)ProfessionalBuilding generative AI applications and agents on AWS (new in 2026)AI path on AWS
Google CloudGenerative AI LeaderFoundationalGenerative AI for business and technology leaders; no codingFor leaders
Google CloudProfessional Machine Learning EngineerProfessionalDesigning, building and running ML on Google CloudML path on Google Cloud
DatabricksGenerative AI Engineer Associate
free prep here
AssociateBuilding generative AI apps on Databricks: RAG, agents, evaluationAI path on Databricks
DatabricksMachine Learning Associate and ProfessionalAssociate, ProfessionalML on Databricks, from first model to productionML path on Databricks
NVIDIAGenerative AI LLMs (NCA-GENL) and Agentic AI (NCP-AAI)Associate, ProfessionalLarge language models; building and running agentsLLM path

Data Engineering

Data engineering certifications

For people who move, clean and shape data so others can trust the numbers. No free prep here yet; each row points to what on this site teaches the same ground.

VendorCertificationLevelWhat it provesLearn for it here
DatabricksData Engineer AssociateAssociateBuilding pipelines on Databricks: Spark, Delta Lake, LakeflowData path on Databricks
DatabricksData Engineer ProfessionalProfessionalProduction data engineering on Databricks: performance, streaming, governanceDataTwin Studio
DatabricksAssociate Developer for Apache SparkAssociateThe Spark DataFrame API and how Spark runs a jobSpark internals, recorded
MicrosoftDP-900: Azure Data FundamentalsFundamentalsCore data ideas and the Azure data servicesKey concepts
MicrosoftDP-700: Fabric Data Engineer AssociateAssociateIngesting, transforming and running data on Microsoft Fabric (DP-203 retired in 2025)Data path on Azure
MicrosoftDP-600: Fabric Analytics Engineer AssociateAssociateModelling and serving data for analysis on FabricMedallion architecture
AWSData Engineer – Associate (DEA-C01)AssociateData pipelines, stores and governance on AWSData path on AWS
Google CloudAssociate Data PractitionerAssociateEntry-level data work on Google CloudData path on Google Cloud
Google CloudProfessional Data EngineerProfessionalDesigning data systems on Google Cloud: BigQuery, Dataflow, Pub/SubData platform architecture
SnowflakeSnowPro Core and SnowPro Advanced: Data EngineerAssociate, AdvancedWorking on Snowflake; then data engineering on itData path, open source
dbt Labsdbt Analytics EngineeringAssociateTransforming data with dbt: models, tests, documentationData contracts
ConfluentCertified Developer for Apache Kafka (CCDAK)AssociateBuilding producers, consumers and streams on KafkaStreaming, explained

Checked 31 Oct 2026. Vendors rename and retire exams often (AWS retired Machine Learning Specialty in March 2026; Microsoft replaced AI-900, AI-102 and DP-100 in June 2026), so confirm on the official page before you book: Microsoft · AWS · Google Cloud · Databricks · NVIDIA · Snowflake · dbt · Confluent.

By vendor

Certifications by vendor

The same exams as above, grouped by who runs them. Other vendors (NVIDIA, Snowflake, dbt Labs, Confluent) are in the tables above.

Claude (Anthropic)

FieldCertificationLevelWhat it provesLearn for it here
AIClaude Certified Architect – Foundations (CCA-F)
free prep here
FoundationalDesigning with Claude: agents, Claude Code, tools and MCP, prompts, contextAgent Harness course
AIClaude Certified Architect – Professional (CCAR-P)
free prep here: guide line by line
ProfessionalThe next level of the same: production agent architectureAgent Harness Series

Databricks

FieldCertificationLevelWhat it provesLearn for it here
AIGenerative AI Engineer Associate
free prep here
AssociateBuilding generative AI apps on Databricks: RAG, agents, evaluationAI path on Databricks
AIMachine Learning Associate and ProfessionalAssociate, ProfessionalML on Databricks, from first model to productionML path on Databricks
DataData Engineer AssociateAssociateBuilding pipelines on Databricks: Spark, Delta Lake, LakeflowData path on Databricks
DataData Engineer ProfessionalProfessionalProduction data engineering on Databricks: performance, streaming, governanceDataTwin Studio
DataAssociate Developer for Apache SparkAssociateThe Spark DataFrame API and how Spark runs a jobSpark internals, recorded

Microsoft

FieldCertificationLevelWhat it provesLearn for it here
AIAI-901: Azure AI FundamentalsFundamentalsAI ideas and the Azure AI services (replaced AI-900 in 2026)The AI Universe
AIAI-103: Developing AI Apps and Agents on Azure
free prep here
AssociateBuilding AI apps and agents on Azure (replaced AI-102 in 2026)AI path on Azure
AIAI-300: MLOps engineeringAssociateTraining, deploying and running models on Azure (replaced DP-100 in 2026)ML path on Azure
DataDP-900: Azure Data FundamentalsFundamentalsCore data ideas and the Azure data servicesKey concepts
DataDP-700: Fabric Data Engineer AssociateAssociateIngesting, transforming and running data on Microsoft Fabric (DP-203 retired in 2025)Data path on Azure
DataDP-600: Fabric Analytics Engineer AssociateAssociateModelling and serving data for analysis on FabricMedallion architecture

AWS

FieldCertificationLevelWhat it provesLearn for it here
AIAWS Certified AI Practitioner (AIF-C01)FoundationalAI, ML and generative AI ideas and the AWS services for themThe AI Universe
AIMachine Learning Engineer – Associate (MLA-C01)AssociateBuilding, deploying and monitoring ML on AWSML path on AWS
AIGenerative AI Developer – Professional (AIP-C01)ProfessionalBuilding generative AI applications and agents on AWS (new in 2026)AI path on AWS
DataData Engineer – Associate (DEA-C01)AssociateData pipelines, stores and governance on AWSData path on AWS

Google Cloud

FieldCertificationLevelWhat it provesLearn for it here
AIGenerative AI LeaderFoundationalGenerative AI for business and technology leaders; no codingFor leaders
AIProfessional Machine Learning EngineerProfessionalDesigning, building and running ML on Google CloudML path on Google Cloud
DataAssociate Data PractitionerAssociateEntry-level data work on Google CloudData path on Google Cloud
DataProfessional Data EngineerProfessionalDesigning data systems on Google Cloud: BigQuery, Dataflow, Pub/SubData platform architecture

Free prep · Anthropic · AI Engineering

Claude Certified Architect – Foundations (CCA-F)

Designing with Claude: agentic architecture, Claude Code, prompt engineering, tool design with MCP, and context management.

Trap gym

The exam's questions are easy to read; the hard part is that every option is a sensible practice. Each trap gym page takes one trap, plays the options out in a simulator, and ends with exam-style practice. First up: Prompt or hook?

Mock exams

Take #7 in plain style first: it reads like the real exam, and all four options work, with one fitting best. Then the original #7 and #6 as a stretch. You can't pass any of them by picking the longest option.

MockQuestionsWhat it is
Mock #7, plain styleStart here60Rewritten after sitting the real exam: the whole situation in plain words, every option a real practice, and each explanation says why the others lose. Mocks #6, #5 and #4 are now in the same style.
Mock #760Exam level: clear questions, close answers. Its replay simulator plays out what each answer would actually do.
Mock #6Stretch60Harder than the exam: every question has a deliberate runner-up.
Mock #560Real-exam style, rebuilt to the same standard.
Mock #460Real-exam style, rebuilt to the same standard.
Mock #260Timed, 120 minutes. Earlier and easier.
Mock #177The full early bank. Earlier and easier.
Mock #3bankQuestions by scenario, for practice rather than timing.

Mind maps

Every question in Mocks #4 to #7 ends with a mind map: a small table of what you see, what is really wrong, and which fix fits. All the mind maps on one page, with similar ones merged and each named by the decision it helps you make. Good for revision the week before the exam.

Study

Free prep · Microsoft · AI Engineering

AI-103: Developing AI Apps and Agents on Azure

The exam that replaced AI-102: planning and managing Azure AI, generative AI and agents, vision, text and document extraction. Most questions are about Azure-specific implementation.

Mock exams

MockQuestionsWhat it is
Mock #1Start here50Two case studies (claims assistant, intake), in the formats the exam uses: single and multiple choice, yes/no, ordering, code and case questions.
Mock #2New50All-new questions with two new case studies (a multilingual support agent, a supplier document pipeline). Take it after Mock #1, timed.

Study

Free prep · Databricks · AI Engineering

Databricks Certified Generative AI Engineer Associate

Building generative AI apps on Databricks: designing the app, preparing data for retrieval, building and deploying agents, governance, and evaluation and monitoring. 45 scored questions in 90 minutes, choose-one and choose-two only.

Mock exams

MockQuestionsWhat it is
Mock #1New45Weighted like the exam across the six sections, with a reason for every option. Opens in practice mode.

Study

Original questions, not real exam content. Databricks has renamed some features since the exam guide (for example Vector Search is now AI Search); the questions use the guide's names and the study pages note the new ones.

Free prep · Databricks · Data Engineering

Databricks Certified Data Engineer Professional

Building, optimizing and maintaining production data pipelines on Databricks: code with Python and SQL, ingestion, Lakeflow Pipelines and Jobs, performance and cost, security, monitoring and deployment with bundles. 60 scored multiple-choice questions in 120 minutes. Built to the official guide dated 9 October 2026.

Mock exams

MockQuestionsWhat it is
Mock #1New60Weighted like the exam across its nine sections. Every wrong option is a working practice with its trap named, and every question has a runner-up. Opens in practice mode.

Study

Original questions, not real exam content. The certification web page lists ten sections; the October 2026 guide has nine, and this prep follows the guide. Several Databricks names have changed (Lakeflow, AUTO CDC, Declarative Automation Bundles, Git folders), so the pages say which name the guide uses.

Next

More free prep

Each new prep gets the same shape: a study guide, practice by domain, and mock exams checked so that no answer gives itself away. Data engineering prep is next in line.