project string | track string | builder string | tools_used list | models dict | infrastructure dict | build_steps list | challenges list | badges_targeted list |
|---|---|---|---|---|---|---|---|---|
Bedtime Story Machine | Thousand Token Wood | ShebMichel | [
"Kiro CLI (AI agent)",
"Python",
"Gradio",
"Modal",
"ffmpeg"
] | {
"story_generation": "nvidia/Nemotron-Mini-4B-Instruct (4B params)",
"illustration": "black-forest-labs/FLUX.1-schnell (12B params)",
"total_params": "~16B"
} | {
"inference": "Modal (Nemotron on T4 GPU)",
"image_gen": "HF Inference API",
"hosting": "Hugging Face Spaces",
"frontend": "Gradio with custom CSS"
} | [
{
"step": 1,
"action": "Designed app architecture - chose Nemotron-4B for text, FLUX.1-schnell for images"
},
{
"step": 2,
"action": "Built Gradio app with storybook UI (custom CSS, alternating scene layout)"
},
{
"step": 3,
"action": "Implemented story generation with structured JSO... | [
"Nemotron-4B not available on any free inference provider - solved with Modal deployment",
"HF token needed fine-grained Inference API permissions",
"JSON parsing reliability from small models - added regex extraction fallback",
"Consistent illustration style - solved with prompt engineering suffix"
] | [
"π© NVIDIA Nemotron Quest (Nemotron-4B model)",
"π Tiny Titan (β€4B story model)",
"π’ Modal Awards (Modal deployment)",
"π Field Notes (blog post)",
"π‘ Sharing is Caring (this trace)"
] |
π Bedtime Story Machine β Agent Build Trace
This dataset contains the build trace for the Bedtime Story Machine project, built for the Build Small Hackathon.
What's in this trace
- Complete build steps from concept to deployment
- Architecture decisions and model choices
- Challenges encountered and solutions
- Tools and infrastructure used
Project
- Downloads last month
- 18