AI Automation Engineer Submission Roadmap - Your Next Step to Certification

To complete your certification, you are required to identify a real-world use case, build an Agentic solution using the Automation Anywhere platform, and document your end-to-end solution. This project demonstrates your ability to think critically about automation opportunities and deliver a working agentic solution.

For support, we will be hosting Office Hours on Wed. June 10, at 7:30am PT / 10:30am ET.

Step 1: Identify a Use Case

• Select a real-world business problem or workflow that would benefit from an agentic solution
• The use case should involve decision-making and dynamic responses, not just linear automation
• Feel free to check out agentic solutions available on the Agentic App Store for inspiration. (Kindly note your solution will be checked against all the listings in the Agentic App store, so please do not submit or reuse the solutions from the Agentic App Store.)

Step 2: Analyze & Justify the Agentic Approach.
Provide a written analysis covering the following in the slide deck template provided:

• Problem statement - what is the challenge or inefficiency being addressed?
• Why an agentic solution? - explain why a traditional rule-based automation is insufficient and how an AI agent adds value
  • This should be graded against the understanding of the role an agent plays and should use language showcased in the Design Framework
• Architecture overview - describe the components, tools, and AI capabilities you plan to use
  • Provide a solution pattern, a justification for the chosen tools. 

    

     

• Expected outcomes - what does success look like? Include measurable benefits where possible (e.g. time saved, error reduction)

Step 3: Build the Agentic Solution

• Build the solution using the Automation Anywhere platform
• The solution must use all of the following components:
  • Mozart Orchestrator
  • AI Agent Builder
  • API Task or Task Bot (justify the usage)(Hint: Don’t use Task Bot just to include REST API calls, but use it to work with local files and UI based apps
  • AI Skill
  • Forms (for Human in the loop)
• Handle error and exception scenarios
• Follow best practices:
  • Follow naming conventions for bots and variables
  • Ensure the solution is environment-agnostic (no hardcoded paths or machine-specific settings)

Step 4: Create & Upload a Video Record a video (3 - 5 minutes) that covers:

• Introduction to the use case and the problem being solved
• Walk-through of your analysis and why an agentic approach was chosen
• Live demonstration of the solution in action
• Summary of outcomes

Upload requirements:

• Upload the video to YouTube
• Ensure the video is clearly titled: "AI Automation Engineer Certification - [Your Name] - [Use Case Name]"
• Share the YouTube link as part of your submission
• Use the following hashtags in the description: #AutomationAnywhere #AIAE

Step 5: Submit Your Documentation. Submit a documentation file covering:

1. Use case analysis - your written analysis from Step 2
2. Solution user guide - step-by-step instructions on how to set up and run the solution, including prerequisites, dependencies, and configuration steps
3. Link to the solution in public folder
4. Test evidence - screenshots or logs confirming the solution ran successfully and the Evals
5. YouTube link - link to your demo video

In summary, you are submitting this template (also linked above) completed, and the link to your Youtube video. You can publish these as “Unlisted” so only people with the link can view.

Evaluation criteria and scoring weight

• Use case relevance & problem statement
• Quality of agentic analysis & justification
• Working Agentic Solution
• Video quality & communication
• Documentation quality & completeness

Submission Process

• Check in your solution to the Public folder. Coming later this week, we’ll share a link where you can share your submission’s URL. 
• Upload the documentation via the template shared above to the CR (in the defined folder structure shown below)

Folder Structure Design:
Create a folder with your solution name under Bots → <your username> (In the screenshot below, consider your name is shown as Capstone Project). It should include the following: (See Screenshot 1)

• If the solution has multiple agents, create a dedicated folder for each agent (See Screenshot 2 below). Each sub-folder should contain:
  • • A Prompts folder - for all AI Skills
    • A Tools folder — for all API Tasks wrapped around AI Skills (See Screenshot 3)
    • The agent itself
• A Documentation folder
• Process and OAuth-related assets can remain in the main folder