Skip to content

Instantly share code, notes, and snippets.

View NoRaincheck's full-sized avatar

NoRaincheck

  • On the Interwebs
View GitHub Profile
name socratic-tutor
description Structured Socratic tutoring when the user explicitly asks to be taught, quizzed, or drilled to mastery on a topic: diagnostic placement, a skill map with levels 0-5, spaced review, and a final assessment. Trigger phrases: 'teach me', 'tutor me', 'quiz me', 'drill me', 'help me master'. Not for one-off 'explain X' or 'what is X' questions.

Socratic Tutor

Act as a personal tutor that takes the user from their current level to genuine mastery of a topic using the maieutic (Socratic) method: teach primarily by asking questions that force the user to reason, predict, and explain — not by lecturing.

@NoRaincheck
NoRaincheck / ticket.py
Created November 5, 2026 04:40
A drop-in replacement for Beads style setup using stdlib Python
#!/usr/bin/env python3
"""ticket.py — a graph-based issue tracker in a single file.
A self-contained, standard-library-only task tracker with dependencies,
parent/child hierarchies, claiming, filters, and SQLite persistence. Run it
directly or import it as a Python module. Python 3.12+ is required.
The file is intentionally self-contained, so it can be copied into a GitHub
gist or downloaded on its own and used as-is. Keep the name ``ticket.py`` when
running the examples below.
@NoRaincheck
NoRaincheck / jev_embeddinggemma.py
Created October 6, 2026 21:52
Jev-like style setup for Embedding Gemma, influenced by mediapipe
import torch
from sentence_transformers import SentenceTransformer, util
import json
# 1. Load EmbeddingGemma (ensure you have a recent version of sentence-transformers)
# Model: https://huggingface.co/google/embeddinggemma-300m or embeddinggemma-2 variants
model = SentenceTransformer("google/embeddinggemma-300m", trust_remote_code=True)
jev_payload = {
"state": "Your app crashed twice during checkout and billed my card twice!",
@NoRaincheck
NoRaincheck / jev.go
Created September 20, 2026 01:00
Golang variant of https://github.com/typesafe-ai/system-one-adapter-python this allows creating a server with the SystemOne endpoints which route to local OpenAI compatible chat completion endpoints.
// Jev-like SystemOne adapter on :1235, backed by LMStudio on :1234.
//
// POST /v1/systemone accepts the TypeSafe request shape
// {state, model, questions} where questions are noul/choice/score, and
// translates each evaluation into an LMStudio OpenAI-compatible chat
// completion with a strict per-request JSON schema. Only models listed by
// LMStudio (GET {base}/api/v1/models, fallback /v1/models) are allowed.
package main
import (

Agentic engineering follows a structured "research, plan, implement" loop. The key focus is distinguishing where a "human leans in" versus where an "agent runs it."

  • 🧠: Human leans in
  • ⚙️: Agent runs it

Research

  1. 🧠🔬 Design Together: Engage with every brainstorming decision and design section yourself. Design is never delegated to agents; delegated design is slop.
    • When Unsure: Seek an adversarial second opinion. A separate agent session on a different model family should judge the decision or design section. (Note: This applies to both brainstorming decisions in Step 1 and spec review in Step 3).
  2. ⚙️📡 Write the Spec: Specs are written for agents, not humans. If you actively engaged in the design phase, you do not need to line-read the document for reassurance.
@NoRaincheck
NoRaincheck / para.md
Created September 14, 2026 09:11
PARA Digital Organization Framework

Project Name: PARA Digital Organization Framework


1. Introduction

This document outlines the structural and operational requirements for implementing the PARA (Projects, Areas, Resources, Archives) file organization system. The objective is to create a universal, action-oriented digital workspace that minimizes decision fatigue, prioritizes active work, and maintains long-term retrievability without relying on complex, rigid hierarchies.

2. Scope

This specification applies to all primary digital workspaces, including local operating system file explorers (Windows, macOS, Linux), cloud storage platforms (Google Drive, Dropbox, OneDrive), and local note-taking applications (Obsidian, Notion, Evernote).

@NoRaincheck
NoRaincheck / ponytail.md
Created September 7, 2026 03:48
Ponytail SKILL. You don't need a massive repo. Source: https://news.ycombinator.com/item?id=49592992
  1. Does this need to exist at all? Speculative need = skip it, say so in one line. (YAGNI)
  2. Already in this codebase? A helper, util, type, or pattern that already lives here → reuse it. Look before you write; re-implementing what's a few files over is the most common slop.
  3. Stdlib does it? Use it.
  4. Native platform feature covers it? <input type="date"> over a picker lib, CSS over JS, DB constraint over app code.
  5. Already-installed dependency solves it? Use it. Never add a new one for what a few lines can do.
  6. Can it be one line? One line.
  7. Only then: the minimum code that works.
  • No unrequested abstractions: no interface with one implementation, no factory for one product, no config for a value that never changes.
  • No boilerplate, no scaffolding "for later", later can scaffold for itself.
@NoRaincheck
NoRaincheck / main.py
Created July 18, 2026 13:22
Chatterbox Voice Cloning - Just use ONNX. Github Instructions are broken for > 1 year
import onnxruntime
from transformers import AutoTokenizer
from huggingface_hub import hf_hub_download, try_to_load_from_cache
import numpy as np
from tqdm import trange
import librosa
import soundfile as sf
MODEL_ID = "ResembleAI/chatterbox-turbo-ONNX"
SAMPLE_RATE = 24000
import argparse
import os
import subprocess
import sys
import tempfile
from pathlib import Path
from PIL import Image
import numpy as np
@NoRaincheck
NoRaincheck / speech_diarizer.py
Created May 22, 2026 12:36
onnxruntime only speech diarizer. Everthing else is essentially stdlib. Based on `pyannote-segmentation`
import urllib.request
import wave
import numpy as np
import onnxruntime as ort
model_id = "onnx-community/pyannote-segmentation-3.0"
model_path = "model.onnx"
urllib.request.urlretrieve(
f"https://huggingface.co/{model_id}/resolve/main/onnx/model.onnx",