Slow & Steady AI：以小步驟減輕壓力｜Hacktoberfest 第 3 週
# Slow & Steady AI: Small Steps, Less Stress | Hacktoberfest Week 3
Slow & Steady AI 是一款輕量級 AI 任務助手，可把大型任務拆成四個較易處理的步驟，協助用戶開始行動。用戶輸入任務並選擇可用時間和精力水平，即可獲得簡單行動計劃。項目使用 Transformers.js 在瀏覽器執行開放權重模型，計劃採用 FLAN-T5 Small，無需付費 AI API 或專用後端。
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady
What I Built
Slow & Steady AI — Small Steps, Less Stress
Slow & Steady AI is a lightweight AI-powered task assistant designed to help students and busy people get things done without feeling overwhelmed.
Sometimes, starting a task is harder than completing it. A big assignment, a personal project, or even a simple to-do list can feel stressful when everything seems to need attention at once.
Slow & Steady AI helps turn a large task into four smaller, manageable steps. Users can enter their task, select their available time and energy level, and receive a simple action plan to help them get started.
The goal is simple: less overthinking, less pressure, and more progress — one small step at a time.
Demo
Live Website: https://ayush6302.github.io/Slow-Steady-AI/
Code
GitHub Repository: https://github.com/AYUSH6302/Slow-Steady-AI
The project is built around a simple web interface using HTML, CSS, and JavaScript.
How I Built It
I designed Slow & Steady AI as a browser-based application using:
- HTML to structure the interface.
- CSS to create a clean, responsive design.
- JavaScript to handle user input, task planning, and progress tracking.
- Transformers.js to run an open-weight language model directly in the browser.
-
FLAN-T5 Small (
Xenova/flan-t5-small) as the planned model for breaking tasks into smaller steps.
The idea is to keep the experience simple and avoid relying on a paid AI API or a dedicated backend. Model inference can run on the user's device, although the model needs to be downloaded initially and browser/device support can affect performance.
Why Does Open Innovation Matter?
Open innovation matters because AI should not be limited to people who can afford expensive APIs or powerful infrastructure.
Open-weight models and open-source tools give developers the freedom to experiment, learn, customize, and build useful applications with fewer barriers. They also make it possible to explore local AI inference instead of automatically sending every task to a third-party service.
For a project like Slow & Steady AI, this approach supports accessibility, experimentation, and greater control over how AI is used.
As a first-year AI/ML student, I see open-source AI as an opportunity to learn by building real projects and make AI tools more approachable for everyday users.
My Agent Session
I have not included an agent session link yet.
The Lazy Payoff
The idea behind Slow & Steady AI is that users should spend less time figuring out where to start and more time taking action.
Instead of staring at a long to-do list, users can turn one overwhelming task into a few clear next steps. The potential benefit is a smoother start, less decision fatigue, and a more manageable workflow.
The goal is not to avoid doing the work — it is to make starting easier.
Prize Categories
Open-Source AI / Local AI Inference
來源：DEV Community · dev.to