bexgboost BexTuychiev

ML Exam Cheat Sheet — Worked Examples

Formulas

Topic	Formula	Variables
Z-score	`z = (x - μ) / σ`	x=value, μ=mean, σ=std dev
Min-max	`x' = (x - xmin) / (xmax - xmin)`	xmin/xmax=dataset min/max
Accuracy	`(TP + TN) / N`	N=total samples
Precision	`TP / (TP + FP)`	"of all predicted positive, how many were right"

Transfer Context

Prepare context for a new chat session when this one is degraded or hitting limits.

File Output

Write the context transfer to a markdown file at .claude/context-transfers/<random-8-chars>.md (relative to the project root). Create the directory if it doesn't exist. Use a random alphanumeric string for the filename.

After writing the file, output ONLY this to the user (nothing else):

Noir Detective RPG - System Prompt

Full system prompt for the Mastra + Mem0 detective agent. Used in the tutorial: How to Build an AI Agent with Mastra and Mem0.

You are the game master for a noir detective text RPG set in 1947 Los Angeles. You narrate scenes, voice NPCs, and track the investigation. Keep the noir tone but write at an accessible reading level. Short sentences. Plain words. A bit of atmosphere goes a long way but don't overdo it. Keep responses to 1-3 short paragraphs. Never more than 150 words per response.

== THE CASE ==
Vincent Moreau, owner of The Blue Note jazz club on Central Avenue, was found dead in his locked office on a Tuesday morning. Cause of death: poisoned whiskey (cyanide in his personal bottle of rye). The office door was locked from the inside. A window opens onto the fire escape. The player is a private detective hired by Claire Moreau, Vincent's sister, who doesn't trust the police to investigate properly.

PUT /memories/{id} Bug Report

Setup

The bug shows up on the self-hosted Docker deployment described in the article. The stack uses three containers:

mem0/mem0-api-server:latest (mem0ai v0.1.117 at time of testing)
ankane/pgvector:v0.5.1
neo4j:5.26.4

	"""Scrape a job board into structured data, then rank roles against a resume.

	Pipeline: scrape the listing, batch-extract each ATS detail page, match to a
	resume with an LLM. One schema works across Ashby, Greenhouse, and any other ATS,
	so there are no per-site selectors to maintain.
	"""

	import json
	import os
	import sys

	"""A multi-step navigation flow against a Hugging Face model page.

	The shape: one scrape opens the live session on the model card. Two further
	interact calls reuse that session to navigate to the Files tab and the Community
	tab and pull data each tab renders. The browser is the same browser across all
	three calls — page state and history carry over.

	Target: Qwen/Qwen2.5-7B-Instruct (popular, public, no gate).
	"""

	description: "Email drafter evaluation"

	prompts:
	- \|
	Draft an email based on these bullet points.
	Match the specified tone throughout the email.

	Bullet points:
	{{bullet_points}}

	"""
	Flask API Authentication Module
	================================
	Monolithic auth module handling token validation, role-based access control,
	and session management. This file is the starting point for a Plan Mode
	refactoring tutorial: splitting it into token_validation.py, role_access.py,
	and session_management.py.

	Repository: https://www.datacamp.com/tutorial/claude-code-plan-mode
	"""

	"""Generate a PDF invoice from client details and line items."""
	import argparse, json, os
	from datetime import date

	from reportlab.lib.pagesizes import letter
	from reportlab.lib.colors import HexColor
	from reportlab.lib.units import inch
	from reportlab.platypus import (
	SimpleDocTemplate, Table, TableStyle,
	Spacer, Paragraph, HRFlowable,

	import asyncio
	import json
	import os
	import sys
	from pathlib import Path

	from dotenv import load_dotenv

	load_dotenv()