Directory / Cisco

Cisco AI Technical Practitioner (AITECH)

A 60-minute conceptual exam on generative AI, prompting, AI-assisted development and agents, including MCP. Despite the Cisco name, the blueprint mentions no Cisco products.

What it signals

Practical assessment: Scenario-based multiple choice. It is a proctored, timed, pass/fail exam. Formats are unpublished and the objectives are knowledge-level (Describe/Explain), so it is at most scenario MCQ, with no labs or code.
Identity assurance: Proctored, arrangements vary by provider. Scheduled through Cisco's CertMetrics portal (official 'Schedule exam' link). English only. Official pages do not say whether delivery is test-centre or online; a third-party guide says Pearson VUE test centre or OnVUE online proctoring (unverified).
Depth: Foundational. Foundational level.
Experience expected: Under 6 months. No prerequisites or experience requirement. It is aimed at working technical staff new to AI.
Portability: Vendor-neutral. The blueprint is vendor-neutral, with no Cisco products. Topics (LLMs, RAG, MCP, prompt injection, governance) apply across any AI stack.
Verifiability: Verifiable on request. Cisco certifications are verifiable through Cisco's certification tracking system and typically through Credly badges, but no public Credly badge for AITECH could be confirmed.

For hiring managers

Despite the Cisco brand, the blueprint is almost entirely vendor-neutral: no Cisco products appear. It covers LLM concepts, RAG, prompting, prompt injection, responsible AI, AI-assisted coding, agents and MCP.
It is a 60-minute conceptual exam with no prerequisites. Cisco's launch blog calls it a 'professional-level certification', but the content is at foundational, AI-literacy-for-technologists depth, far from a CCNP-style professional exam. It does not prove someone can build or ship an LLM app.
It is notable for giving agentic AI 20% of the weight, including MCP primitives and human-in-the-loop design. That is broader agent and MCP coverage than most entry-level vendor certifications.
Sibling path: the AI Business Practitioner (AIBIZ) is a non-technical Cisco U. learning path for business professionals and managers (genAI essentials, prompt and context engineering, responsible adoption, AI strategy, pilot projects). It ends in a Cisco badge, not a proctored exam, so it is weaker evidence than AITECH. Cisco positions AIBIZ, then AITECH, as a progression.
Cisco ran a 'Rev Up to Recert: GenAI' promotion (2026-02-09 to 2026-03-26) with free access to the AITECH and AIBIZ paths, so a wave of holders certified in spring 2026.

Verifying it

Cisco certifications are tracked in Cisco's certification tracking system (CertMetrics), and Cisco issues many credentials as Credly badges. No official Credly badge page for AITECH was found (only other Cisco AI learning badges on Credly). A LinkedIn profile listing 'Cisco Certified AI Technical Practitioner' suggests shareable credentials exist. Employers should ask for a Credly link or a Cisco verification record.

What it covers

Published exam outline

Generative AI Models 20%
Prompt Engineering 15%
Ethics and Security 15%
Data Research and Analysis 10%
Development and Workflow Automation 20%
Agentic AI 20%

By topic

GenAI apps: Substantial (20 to 35%). Generative AI Models (20%, including RAG, embeddings, vector DBs, model selection, context windows) plus the non-security parts of Prompt Engineering (~11%) and token/cost items in domain 5 total roughly 30-35%. This is at the conceptual level only.
Agents: Substantial (20 to 35%). The Agentic AI domain is 20% (agent design, orchestration, MCP primitives, HITL). With workflow-design objective 5.3 the total is roughly 20-25%.
Classical ML: Not covered. There are no model-training objectives. Fine-tuning vs RAG appears in the training course but not in the exam topics.
Operations: Touched on (under 10%). Only cloud vs local hosting trade-offs (1.2), workflow monitoring principles (5.3) and the deployment phase within SDLC (5.1) apply, under 10%.
Data: Some (10 to 20%). AI-assisted EDA, automated data preparation and data privacy controls (4.1-4.3) plus data transformation within agents (6.5) total roughly 10-12%.
Responsible AI: Substantial (20 to 35%). Ethics and Security is 15%. Add prompt injection (2.3), data-exposure controls (4.3) and HITL (6.4) and the total is roughly 20-25%.

Exam details

Who the provider says it is for. Technical professionals adopting AI in their work. Official roles: IT and network engineers, data analysts, AIOps specialists, solutions architects, technical leads, managers and business process analysts. Potential roles named by Cisco: AI engineer, prompt engineer, AI workflow coordinator.

Format: Not published for this exam. Every blueprint objective uses knowledge verbs (Describe, Explain, Compare, Differentiate, Understand), which points to knowledge-based multiple-choice-style items. No labs or simulations are mentioned.
Delivery: Scheduled through Cisco's CertMetrics portal (official 'Schedule exam' link). English only. Official pages do not say whether delivery is test-centre or online; a third-party guide says Pearson VUE test centre or OnVUE online proctoring (unverified).
Code or hands-on work: No coding or labs on the exam as far as published; objectives are conceptual. The Cisco U. learning path advertises hands-on labs, but these are preparation, not assessment.
Prerequisites: None (no prerequisites listed for the certification or the training)
Recommended experience: None stated. The training targets technical professionals 'seeking to transition from traditional knowledge-based work to innovation-driven roles augmented by AI'.
Renewal: Renew before expiry by earning Continuing Education (CE) credits or by retaking exams (official certification page). The number of CE credits needed for AITECH was not found officially. The AITECH Cisco U. training itself awards 8 CE credits. The official exam page says 'This exam can be used toward recertification' (that is, of other Cisco certifications). One search snippet of Cisco's recertification page appeared to say AITECH requires retaking '820-605 CSM' with no CE option. That looks like a copy error from the Customer Success Manager entry and could not be confirmed because the page blocked access.

Preparing

Not confirmed

These details were not available from an official source at the time of research.

questionCount (not published; ~55-65 is a third-party claim)
question formats (not published)
delivery mode (Pearson VUE test centre / OnVUE online claimed only by a third party)
exact general-availability date (Cisco said December 2025)
existence of a Credly badge for AITECH
number of CE credits required to recertify AITECH; conflicting snippet on Cisco's recertification page referencing 820-605 CSM
exam code: some Cisco-hosted pages and Global Knowledge show '800-110', but the current official pages and blueprint use 810-110
current AIBIZ price and whether its badge is issued on Credly

Sources (15)