# Potencializando Seus Estudos e Carreira com IA: A Three-Level Framework for AI-Assisted Learning > This DIO course repository teaches technology career changers how to use AI tools across three levels of interaction: chatbots for advice, copilots for document work, and agents for file system automation. The guide is practical, opinionated, and deliberately tool-agnostic, because the authors expect the tooling landscape to shift faster than the underlying concepts. **digitalinnovationone/potencializando-estudos-carreira-com-ia** — Curso "Potencializando Seus Estudos e Carreira com IA: De Chatbots a Agentes" - Repository: https://github.com/digitalinnovationone/potencializando-estudos-carreira-com-ia - Stars: 4,939 · Forks: 219 - Language: Unknown - License: not declared - Published: 2026-09-23 · Updated: 2026-09-23 · Language: en - Canonical page: https://hysenlabs.com/projects/digitalinnovationone-potencializando-estudos-carreira-com-ia ## Who this course is for and what problem it addresses The repository from Digital Innovation One targets people who are making a career transition into technology and have little or no prior experience with AI tools. The central problem it addresses is decision fatigue: there are dozens of AI tools available, they change rapidly, and a newcomer cannot tell whether they are using a chatbot, a copilot, or an agent, or why the distinction matters. The guide does not assume programming knowledge or any specific tool subscription. It opens by stating that each lesson starts from zero on its topic, and the course as a whole progresses through three levels of autonomy. Level one involves asking and receiving. Level two involves a tool observing your work and suggesting while you act. Level three involves handing a task over and reviewing the result. The README explicitly says that this conceptual hierarchy is more durable than any product recommendation, because tools appear and disappear constantly. A reader who understands the three-level model can evaluate the next wave of products on their own terms. That is an unusual and defensible editorial choice: the authors are teaching a mental model, not a workflow tied to a specific vendor. ## Level one: using a chatbot to choose a tech area The first practical exercise asks the learner to open any chatbot, such as ChatGPT or Claude, and use a structured prompt to identify which area of technology fits their background. The README provides the exact prompt to use: ``` Quero entrar na área de tecnologia, estou em transição de carreira e tenho 1h por dia para estudar. Meu background é em [sua área de origem, ex.: Psicologia]. Me ajude a decidir por onde começar. Responda assim: 1. Três áreas que combinam com meu perfil, em ordem de afinidade 2. Uma sugestão final de qual escolher e o porquê 3. Cinco tópicos iniciais para estudar nessa área Use linguagem simples, como se fosse uma conversa com um amigo. ``` The learner is instructed to copy the response, specifically the chosen area and the five initial topics, into a notes document. That output becomes the input for the next exercise. The chatbot level is described as consulting a specialist available 24 hours a day: you ask, it responds. The guide is clear that the quality of the answer depends directly on the quality of the question. There is no pretense that the chatbot's output is authoritative. The exercise is about extracting a starting direction, not a final answer. ## Level two: turning a chat output into a structured plan with a copilot The second exercise moves from conversation to document editing. The learner opens a blank Google Docs file, activates the integrated Gemini copilot, and uses a second prompt that references the output from the chatbot exercise. The guide distinguishes the copilot from the chatbot in a specific way. The chatbot knows only what you type. The copilot can access the context of the application it is embedded in. In Google Docs with Gemini, that context includes Drive files and Gmail messages. The README prompt instructs the copilot to consider content from the learner's own emails and Drive when suggesting study materials. After the copilot generates an initial draft, the learner refines it in two steps. First, they edit the document directly while the copilot offers real-time continuations. Second, they select a paragraph and ask the copilot to rewrite it in a clearer, more motivating style: ``` Reescreva este trecho de forma mais clara e motivadora. ``` The finished document, a 30-day study plan with weekly topics and resources, becomes the input for the third exercise. The guide points out that this is where the copilot's advantage over the chatbot becomes visible: the suggestions are shaped by materials the learner has already encountered. ## Level three: materializing the plan as a file system with an agent The third exercise uses Google Antigravity as the agent environment. The learner creates an empty folder on their desktop, opens the agent pointing at that folder, and pastes the content of their study plan document into a structured prompt. The agent is asked to create a README, a chronogram file, a folder for notes with a template for the first week, and a resources file. The prompt instructs the agent to explain its structure before executing, giving the learner a chance to review the plan before any files are created: ``` Segue o plano de estudos que escrevi com a ajuda de um copiloto: --- [cole o conteúdo do documento da prática anterior] --- Materialize esse plano nesta pasta criando: 1. README.md com o resumo do plano 2. cronograma.md com o roteiro semanal (semanas 1 a 4) 3. uma pasta "anotacoes" com um modelo "semana-01.md" para eu duplicar 4. recursos.md com os materiais separados por semana Antes de criar, me explique brevemente sua estrutura. Depois execute. ``` The README emphasizes that the review step is where the learning happens. The learner reads each file, notices decisions they would make differently, and adjusts. The guide frames the agent not as a replacement for judgment but as a mechanism that produces a draft the learner can critique. That framing is intentional: the course positions the human as the validator, not the passenger. ## The tool ecosystem and why the guide avoids endorsing specific products The README includes a tools section organized by level: chatbots, copilots, and agents. The README does not list every tool in that section, but the guide's own framing explains its approach. The authors state in a highlighted note that the AI world changes at an accelerated pace and tools appear, disappear, and transform constantly. The recommendation is to understand the concepts first. This is a meaningful constraint on the repository's usefulness. If a learner follows the specific tools listed today and returns in six months, some of those tools may have changed their pricing, their feature set, or their availability. The guide acknowledges this directly and asks learners to follow the repository itself rather than treating the tool list as permanent. For a course aimed at career changers who may not have the technical background to evaluate new tools independently, this creates a real gap. The conceptual framework survives the tool churn, but the practical exercises depend on specific tools that may or may not still exist in the same form. The Google Docs and Google Antigravity exercises will break if those tools change their copilot behavior. ## What the course does not cover The guide does not cover programming. There is no code to write, no environment to configure, and no package to install. It is a conceptual and prompt-engineering introduction, not a software development course. It does not cover any AI model in technical depth. The prompt templates are conversational, not designed to exploit specific model capabilities. A learner who completes the three exercises will know how to describe a request at each level of AI interaction, but will not know how a language model processes that request or how to fine-tune one. The guide also does not address evaluation. There is no rubric for assessing whether the chatbot's area recommendation was good, whether the copilot's study plan was realistic, or whether the agent's file structure was well-organized. The exercises produce outputs, but the course leaves assessment entirely to the learner. Finally, the repository has no graded assessments, no certificate path, and no community forum baked in. Digital Innovation One offers those elsewhere, but this specific repository is a standalone README-driven guide. ## Maintenance and license The last push to this repository was on 2026-05-05. The repository is not archived and remains accessible. The README promises periodic updates as the tooling landscape changes, but it does not commit to a specific schedule. No license file is listed in the repository structure, and the license field is listed as unknown. Reusing or redistributing the prompt templates or course structure without permission carries legal uncertainty. The prompt templates are presented as open for learners to use in AI tools, but any commercial reuse of the repository contents should be confirmed with the copyright holder. The repository has no GitHub Releases, so there is no versioned snapshot of the guide. If the content changes, there is no stable reference point to return to. For a course that explicitly expects the content to evolve, this is a practical limitation for educators who want to teach from a fixed version. ## Conclusion This repository is the right starting point for someone in Brazil entering tech who wants a structured mental model for working with AI tools, not a product tutorial. It is not the right resource for someone who needs runnable software, a graded course, or coverage of any specific AI platform in depth. Before committing, check whether the recommended tools section has been updated recently, because the repository explicitly warns that the tooling landscape changes faster than the guide. The last push was on 2026-05-05, so verify that the linked tools still exist and match their descriptions. ## FAQ ### Does this course require prior programming or AI experience? The README states explicitly that each lesson starts from zero on its topic, and the course is aimed at people in career transition with no requirement for prior AI experience. No software installation is needed for the first two exercises. ### Which AI tools does the guide specifically recommend for the agent exercise? The README names Google Antigravity as the agent environment for the third exercise. The chatbot and copilot exercises are described as compatible with any tool of the learner's preference, with ChatGPT and Google Docs with Gemini given as examples. ### Can the three-level framework be applied to learning goals outside technology careers? The README presents the chatbot, copilot, and agent framework as a general mental model for AI interaction. The exercises are designed around a technology career plan, but the README does not restrict the framework to that domain. ## Sources - [digitalinnovationone/potencializando-estudos-carreira-com-ia on GitHub](https://github.com/digitalinnovationone/potencializando-estudos-carreira-com-ia) - [Issues](https://github.com/digitalinnovationone/potencializando-estudos-carreira-com-ia/issues) - [README](https://github.com/digitalinnovationone/potencializando-estudos-carreira-com-ia/blob/main/README.md) --- Hysen Labs editorial analysis, written from the project's own repository and release notes. Cite the canonical page: https://hysenlabs.com/projects/digitalinnovationone-potencializando-estudos-carreira-com-ia