01
From experiments to systems
Production AI demands reproducibility: versioned data, tracked experiments, validated deployments, and measurable outcomes—beyond a notebook that “works once.”
GIofAI Professional Certification
Certified AI Engineer (CAIE)™
The exam validates a candidate’s ability to apply AI engineering foundations, data engineering, model development, deployment/MLOps, governance, and applied use cases to real-world scenarios. It is designed for professionals who have completed the GIofAI AI Engineering learning pathway or have equivalent industry experience.
Secure proctored assessment
Vendor-neutral capability validation
Verifiable certificate & transcript
Professional designation
CAIE™ Career level
Mid Level
CAIE™ Career level
Mid Level
Duration90 minutes
Questions250 MCQs
Pass threshold70%
LevelIntermediate to Advanced
DeliveryGIofAI Exam Portal
Self-paced Course
Learn with labs & projects at your own pace.
Practice Exam
Timed questions with instant feedback & review.
Professional outcome
AI systems are shipping faster—engineering rigor must keep up.
As AI moves from prototypes to production, teams need repeatable practices across data, model training, deployment, monitoring, and governance. CAIE focuses on the engineering realities that determine whether AI succeeds in the real world.
02
Monitoring is now mandatory
Drift, data quality changes, and shifting user behavior can quietly break models. Engineers need robust observability, alerting, and retraining strategies.
03
Governance meets engineering
High-impact AI requires security controls, documentation, audits, and responsible practices—implemented as engineering workflows, not last-minute checklists.
Who it suits
Designed for professionals working where capability meets real delivery.
Designed for professionals applying these capabilities in real delivery environments.
Practitioners building production systems
For professionals shipping models, pipelines and AI-enabled products.
Engineers & technical specialists
For people responsible for development, evaluation, deployment and monitoring.
Product & solution professionals
For teams translating requirements into reliable, responsible AI solutions.
Consultants & technical strategists
For advisors guiding architecture, adoption, risk and implementation choices.
Exam blueprint
6 domains that reflect the full assessment scope.
The assessment spans the domains below. Weightings indicate relative emphasis in the exam.
Foundations & Data for AI Engineering
20%AI vs ML vs DL, lifecycle, and engineering roles Core math/stat: optimization, loss, regularization Framework concepts: PyTorch, TensorFlow, JAX, ONNX Reproducibility: versioned ar...
Model Development & Training
25%Classical ML: trees, ensembles, clustering, PCA DL: CNNs, RNNs, Transformers, hybrid approaches GenAI: VAEs, GANs, diffusion, LLM concepts Tuning: optimizers, schedules, AutoML app...
Deployment, MLOps & AI Infrastructure
20%Batch vs real-time inference; edge vs cloud/hybrid Serving: REST/gRPC, containers, serverless concepts Enterprise integration: APIs, microservices, vector DBs RAG pipeline componen...
Evaluation & Monitoring
15%Metrics: classification, regression, NLP, vision Robustness: stress tests, shifts, reproducibility Monitoring: drift, dashboards, logging, alerting Online evaluation: A/B tests and...
AI Governance, Security & Compliance
10%Risk frameworks: NIST AI RMF, ISO/IEC 42001 (conceptual) FATE: fairness, accountability, transparency, explainability Security: poisoning, inversion, prompt injection, API hardenin...
Applied AI Engineering Case Studies
10%Retail: recommendations, vector search, bias trade-offs Finance: fraud, explainability, high-risk governance Healthcare: safety, human-in-the-loop, PHI constraints Manufacturing &...
Assessment experience
Know the format before you commit.
A 90-minute assessment with 250 questions, a 70% passing threshold, delivered through the GIofAI Exam Portal.
Exam at a glance
Duration90 min
Questions250
FormatMultiple Choice
Pass70%
Negative markingNone
LanguagesEnglish
Your certification journey
1
Prepare against the published blueprint
Use official study materials and the practice assessment to identify weaker domains.
2
Complete identity verification
Government ID and candidate verification are required for the professional assessment.
3
Sit the proctored examination
Complete the assessment through the GIofAI Exam Portal under monitored conditions.
4
Receive your provisional result
See the provisional outcome immediately after submission.
5
Access your verified credential
Your certificate and detailed transcript are available in the learner dashboard after verification.
Credential value
A designation you can verify, share and carry into professional conversations.
The value of the certification is the combination of controlled assessment, identity verification, domain-level performance evidence and a verifiable professional credential.
CAIE
Certified AI Engineer
A professional GIofAI designation supported by a digital certificate and official transcript. The transcript records section-level performance, giving the credential more context than a simple completion badge.
01
Verified professional designation
Identity-verified assessment under controlled examination conditions.
02
Detailed performance transcript
Section-wise performance information for a clearer view of your assessed capability.
03
Shareable digital certificate
A professional credential designed to be verifiable and shareable with employers and professional networks.
04
Clear next-step pathway
Use CAIE as a core credential before moving into deeper specialisations.
CAIE Exam Guide
Review the blueprint before you register.
Get the certification overview, domain weightings, exam format and preparation guidance in one concise guide.
Frequently Asked Questions
Details for candidates and teams evaluating the credential.
Who is CAIE designed for?
CAIE is built for professionals working or aspiring to work as AI Engineers, ML Engineers, MLOps Engineers, or AI-focused Data Engineers—especially those building production systems.
What is the assessment format?
The assessment is online and remotely proctored, timed at 120 minutes, and uses single-correct MCQs including scenario-based items that mirror real AI engineering decisions.
Is it open-book?
 No. The assessment is closed-book with no external assistance, to preserve the integrity of the credential.
What topics are emphasized most?
The highest weight is on Model Development & Training, plus a strong focus on Foundations & Data Engineering and Deployment/MLOps. Evaluation, monitoring, governance, and applied case studies complete the blueprint.
What do I receive after passing?
You receive a certificate-style credential with an ID and issue date, plus a transcript-style view that can summarize performance signals for verification workflows.
Prove you can ship AI systems—reliably, responsibly, and at scale.
CAIE validates engineering-grade competence across data pipelines, model development, deployment/MLOps, monitoring, and governance—through a structured, scenario-led assessment experience.