AWS Builder Center
Weekend Creative Agent Challenge: Live Goal Coach AI

Weekend Creative Agent Challenge: Live Goal Coach AI

Live Goal Coach AI is an autonomous productivity companion that tracks goals and proactively generates personalized coaching, challenges, and next actions using Amazon Bedrock and AWS Lambda. Scheduled automation allows the AI to work in the background, so users can return to fresh guidance without manually triggering it. Your goals keep moving, even when you're away.

Vision & What It Does

Most goal-tracking applications simply show users their progress and wait for them to decide what to do next. Live Goal Coach AI takes a different approach by turning goal tracking into an always-on AI coaching experience.
The vision is simple: your productivity coach shoul

d work even when you are not looking at it.
A user can create a goal such as “Complete My AI Project” with a description such as “Build and deploy my AI application.” The application stores the goal and tracks its progress. Instead of requiring the user to manually request assistance, the autonomous agent periodically generates fresh, personalized coaching based on the goal and its current state.
The creative output can include motivational messages, creative challenges, suggested next actions, mini action plans, and milestone recommendations. For example, instead of displaying only “Progress: 40%,” the agent can generate: “You are almost halfway through your journey. Today's challenge is to spend 30 focused minutes completing one unfinished module of your AI project.”
This makes the application more than a dashboard. It becomes an always-on productivity companion that proactively prepares useful guidance for the user.

How I Built It

I started by building a simple goal-management web application using HTML, CSS, and JavaScript for the frontend and Python with Flask for the backend.
The initial application allowed users to create goals, add descriptions, view goals, update progress, and manage their goal information through REST API endpoints.
The main challenge was making the application genuinely autonomous. I did not want the AI feature to be another button that required the user to manually click “Generate Coaching.” That would make it an AI-enabled application, but not an always-on agent.
To solve this, I designed a scheduled workflow where an AWS event automatically triggers the coaching process. The workflow retrieves the goal and its current progress, sends relevant context to Amazon Bedrock, generates creative coaching content, and stores the result for the user.
This approach allows the AI to perform useful work independently and have the result ready before the user returns to the application.

AWS Services Used / Architecture Overview

The main AWS service used for the AI workflow is Amazon Bedrock, which provides the generative AI capability for creating personalized coaching content.
AWS Lambda executes the autonomous coaching logic. A scheduled event triggers Lambda without requiring manual interaction from the user.
The architecture is:
Scheduled Trigger
↓
AWS Lambda
↓
Read Goal + Progress
↓
Amazon Bedrock
↓
Generate Creative Coaching
↓
Store Result
↓
Live Goal Coach Dashboard

The application therefore combines a traditional web application with a scheduled generative-AI workflow.
The core idea is:
Schedule → Agent → AI Generation → Stored Output → User
This demonstrates autonomous, scheduled functionality because creative content is generated without the user initiating every generation request.

What I Learned

This challenge helped me understand that building an AI application is not only about connecting an application to a language model. The most important part is designing the complete workflow around the AI.
I learned how scheduled cloud execution can be combined with generative AI to create an application that performs useful work independently. I also learned more about integrating Amazon Bedrock into an application workflow and designing prompts around structured user context such as goals and progress.
Another important lesson was the difference between an AI feature and an AI agent. A chatbot that responds after a user clicks a button is interactive, but an autonomous workflow can initiate useful work based on an event or schedule.

Link to App or Repository

Live Application / Public GitHub Repository:  https://github.com/naveenhari2005/LiveGoalCoach 

Conclusion

Live Goal Coach AI transforms passive goal tracking into proactive AI-powered coaching. By combining goal management, generative AI, scheduled automation, AWS Lambda, and Amazon Bedrock, the application can create personalized creative guidance without requiring continuous manual interaction.
The goal is not simply to tell users where they are.
It helps them decide where to go next - even when they are away.
Live Goal Coach AI - Your goals keep moving, even when you're away.
Any opinions in this article are those of the individual author and may not reflect the opinions of AWS.
Enjoyed reading this content? Let the author know!

Your likes, comments, shares, and saves help creators reach more builders.

Loading recommendations

Loading article