Posted on by Colin Smythe

An Update on Our Navajo Workforce Services Web-app

In October 2025 we launched our Navajo Workforce Services web-app. It has been a busy year.

There has been an explosion of new LLM releases in Generative-AI. Each release has a significant effect on solutions using them in terms of: establishing a sustainable business model; the ways in which the solution is architected; and, more significantly, the quality and quantity of the information being returned in responses to the prompt engineering (and the corresponding human-in-the-loop experience). While this is great when using AI-enabled tools (such as Claude Code to support software development) it is a nightmare when creating an AI-enabled tool and needing to complete extensive testing to create a robust, reliable and stable web-app solution.

We have been working on the use of Generative-AI in two related areas:

    • A new web-app that supports learners, teachers/faculty, mentors and employers to create and use training/learning digital content that is personalized to the learner so they can acquire the specific knowledge, skills and abilities they need for their career development. This includes development of the material using well established pedagogical methods based on concept maps and reinforced learning. More about this in August (2026) when we hope to make the web-app available for early adopters
    • A revamp of the Workforce Services web-app to make it easier to support, to extend its functionality and to simplify integration with any third-party job board. The original web-app was specific to job seekers and employers of the Navajo Nation*.

The current version of the Navajo Workforce Services (https://navajo-workforce.edtechforlearning.org) web-app is designed to support two user scenarios. The first enables job seekers to search and identify suitable job opportunities and the second enables employers to create clear job descriptions so that job seekers can understand their suitability for those jobs: at present the operational version of the web-app only supports job seekers. These scenarios draw upon three key features: the job seeker provides their resume (as a PDF); the employer makes their job descriptions available through the corresponding job board; Generative-AI is used to identify the Knowledge, Skills and Abilities (KSAs)** required by the job description, attained by the job seeker, and the matching between the job descriptions and the resume.

From the job seeker’s perspective, the user story is very simple (a short video of the workflow is available at (https://edtechforlearning.org/workforce-services-app/):

    1. You login to the web-app (https://navajo-workforce.edtechforlearning.org) using your Google account
    2. Next, use the ‘For job Seekers’ menu link after which various options for providing your job resume are presented. On the first visit you upload a PDF version of your resume (typically 1-2 pages long but the web-app will accept any length document).
    3. To activate the upload you click on the ‘Perform job match’ button. The resume is now uploaded and the web-app uses Generative-AI to inspect the resume and identify the set of Competency Statements (each of which consists of the corresponding set of KSAs)
    4. Once this automated analysis is completed the list of competency statements is presented. The details of the statements can be explored and refined by clicking on the ‘Suggested Changes’ button alongside each statement. This ‘Human-in-the-Loop’ chat process is moderated using Generative-AI and allows the job seeker to confirm/improve the accuracy of the statement
    5. The next step, activated by clicking the “next’ button, is to either compare your competency statements against the set of available job descriptions or to save the statements for later use. The list job descriptions will have been uploaded/updated and analysed previously by the web-app accessing the corresponding job board
    6. The ‘Match Against All Available Jobs’ button is used to start the search against the set of job descriptions available to the web-app. Upon completion of the search the list of those job descriptions with the best match (an algorithm that requires a match correlation of at least 75%) is presented. This list provides access to the full job description and also identifies the competencies that are missing from the job seekers resume.
    7. The final step is to complete the process by leaving the web-app and to receive the list of saved KSAs (these can also be accessed through later uses of the web-app by the job seeker).

Once we have sufficient usage and feedback from job seekers we can improve their user experience and then make available the employers’ capabilities. For employers, the web-app enables them to inspect and amend the analysis of the job descriptions by Generative-AI to confirm/improve the accuracy of the KSAs associated with the job. We will then improve the comparison between the job seeker’s KSAs and those required by the set of available job descriptions to provide detailed skills-gap analysis followed by recommendation of the appropriate available digital learning content. More details on these extended capabilities and the full ‘Achievements Pipeline’ are available in our White Paper (https://edtechforlearning.org/projects-achievement-pipeline/). At a more general scope we will improve support for:

    • Accessibility (both in terms of the user experience of the web-app as a whole and, perhaps more significantly, the interaction with the various chat-bots that provide the human-in-the loop interactions). We will use open standards to enable the appropriate configuration of the web-app user experience
    • The collection of usage data so that we can understand how the web-app is being used and the strengths and weaknesses of Generative-AI for both the creation of the KSAs and the human-in-the-loop chatbots
    • A user to provide feedback on their experience and to suggest new functionality.

If you work with a community that provides job boards and want to make tools available to support use of a job board, or know someone who does, then we would love an introduction.

* This work was supported by a grant from GitLab Foundation (https://www.gitlabfoundation.org) under their AI for Economic Opportunity (https://www.gitlabfoundation.org/our-journey/ai-economic-opportunity-fund-inaugural-grantees) programme and we thank them for support of our work.

** Knowledge is ‘what you know’ in terms of concepts, principles and procedures. Skills are the learned proficiencies developed through the practice of the application of the knowledge. Abilities are the innate capabilities and the capacity acquired through experience to perform a skill.

Also published on the EdTech For Learning LinkedIn page: https://www.linkedin.com/pulse/update-our-navajo-workforce-services-web-app-edtech-for-learning-wp5xe/