Nicheloom

Market intelligence for builders — see what's gaining traction before it's crowded.

I built a local AI-powered Ouija board with a fine-tuned 3B model

Details

External ID
47169979
Source
HN
Company
—
Product
I built a local AI-powered Ouija board with a fine-tuned 3B model
Website domain
github.com
Launched
Feb. 26, 2026
Cohort
—
Upvotes
6
Upvotes percentile
0.28099730458221023
Tags
—
Fetched at
Sept. 7, 2026, 9:25 p.m.
Updated at
Sept. 7, 2026, 9:25 p.m.

Enrichment

Theme
Vertical
Horizontal
Function
Hardware & robotics
Audience
B2C
AI stance
AI-native
Project type
Hobby / open-source project
Normalized one-liner
ai-powered ouija board
Manually corrected
False

Could you build this?

Partial While prompting a local model via llama.cpp or Ollama is trivial, fine-tuning an open-weights 3B model and integrating it with specific hardware or interactive interfaces requires domain-specific ML workflow setup.

What it would actually take: Requires curating a stylized dataset, setting up LoRA/QLoRA fine-tuning pipelines using PyTorch/Unsloth, quantizing weights to GGUF format, and interfacing with a local inference engine. While AI can write the glue code, preparing training datasets and tuning hyper-parameters for coherent behavioral styling requires hands-on ML experimentation.

Discussion

5 comments analyzed.

Concerns raised: Real crisis detection needed for sensitive content

Feature requests: iOS version, UX feedback on board animations and planchette movement, Writeup/documentation of fine-tuning methodology

Competitors

Other products that read as similar to this one — 231 launches clear the similarity bar, closest 8 shown.

Attention rank: #170 of 232 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).

Launched 120 days after the earliest competitor.

Other launches for this product

Same idea, different domain

Nobody's really built a hardware & robotics tool for Fintech yet.