AWS Builder Center

Weekend Creative Challenge: Roast My Bio

Have you ever read a LinkedIn bio and thought — this person takes themselves way too seriously? I have. That's exactly what inspired me to build Roast My Bio, an AI-powered web app that reads your professional bio and roasts it in three funny sentences. The goal was simple: make something that makes people laugh, runs entirely on AWS, and can be built in a single morning.

Weekend Creative Challenge: Roast My Bio

Every weekend challenge needs a spark. Mine came from scrolling through LinkedIn and reading bio after bio filled with words like "passionate", "synergy", and "thought leader". I thought — what if AI could roast this? That became Roast My Bio. A simple, fun app that uses Amazon Bedrock to generate a 3-sentence comedic roast of any professional bio. Built in one morning. Deployed entirely on AWS Free Tier. Zero dollars spent.


Vision & What The App Does

Roast My Bio is an AI-powered web app that takes your LinkedIn or Twitter bio and roasts it in 3 funny sentences. The idea is simple: professional bios are often overly serious and full of buzzwords. This app uses AI to poke fun at that in a playful, lighthearted way.
The creative output is a short comedic roast generated instantly. Users paste their bio, click a button, and get roasted by an AI comedian. It's fun, shareable, and makes people smile — which is exactly the point of this challenge.


How I Built It

I built this over a single morning using AWS free tier services only.
The frontend is a single HTML file with vanilla JavaScript — no frameworks, no build tools. I kept it intentionally simple so the focus stays on the creative output rather than the UI complexity.
The backend is an AWS Lambda function written in Python 3.12. It receives the bio from the frontend, builds a prompt for the AI, calls Amazon Bedrock Nova Lite, and returns the roast as a JSON response.
The biggest challenge was getting the Bedrock Nova Lite API request format right. The content field needs to be a JSONArray, not a plain string, and maxTokens lives inside inferenceConfig — not at the top level. These were small but important details that took some debugging to get right.
I used API Gateway as the bridge between the frontend and Lambda, with CORS configured in two places — API Gateway settings and inside the Lambda response headers. Missing either one causes silent failures in the browser with no useful error message.

AWS Services Used / Architecture Overview

Service
Purpose
Amazon S3
Hosts the static frontend (index.html)
AWS Lambda
Runs the backend Python code
Amazon API Gateway
HTTP API connecting frontend to Lambda
Amazon Bedrock (Nova Lite)
Generates the AI roast
Service
Purpose
Amazon S3
Hosts the static frontend
AWS Lambda
Runs the backend Python code
Amazon API Gateway
HTTP API connecting frontend to Lambda
Amazon Bedrock Nova Lite
Generates the AI roast

Architecture flow:
User Browser → S3 Static Website → API Gateway (POST /roast) → Lambda (Python 3.12) → Bedrock Nova Lite → Returns roast
All services run on AWS Free Tier. Total cost for building and testing: effectively $0.


What I Learned

  • Bedrock Nova Lite API format is specific — the messages content must be a JSONArray and inference parameters go inside inferenceConfig, not at the root level. This caused validation errors that took debugging to identify.
  • S3 static website hosting is underrated — upload one file, enable hosting, add a bucket policy and your frontend is live globally in under five minutes. No web server, no maintenance.
  • HTTP API vs REST API in API Gateway — for a single POST route, HTTP API is faster to set up, cheaper to run, and has built in CORS configuration. No reason to use REST API for simple use cases.
  • CORS needs two places — API Gateway handles the preflight OPTIONS request. Lambda response headers handle the actual POST response. Both are required or the frontend breaks silently.
  • Lambda plus Bedrock is powerful — no servers to manage, auto scaling built in, and you can build a working AI app in under two hours. For weekend projects this stack is hard to beat.


Conclusion

Roast My Bio started as a joke idea and turned into a genuinely fun app that people actually want to try. The challenge pushed me to learn how Bedrock Nova Lite works, how to wire together Lambda and API Gateway properly, and how fast you can go from idea to deployed product on AWS.
If you have a weekend and an AWS account, you have everything you need to build something creative. Start simple, deploy early, and don't overthink it. The jacket doesn't hurt either.


Link To App

Live app:  live app link 
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