zseta’s gists

zseta / semantic_cache.py

Created November 5, 2025 19:07

	def semantic_cached_prompt(prompt):
	"""Retrieve a response from ScyllaDB or ask OpenAI if it's a new prompt.
	Args:
	prompt (str): The user prompt.
	Returns:
	str: The response to the prompt.
	"""
	embedding = create_embedding(prompt)
	cached_response = search_cache(embedding, threshold=0.80)
	if cached_response:

zseta / search.py

Created November 5, 2025 18:54

	def calc_cosine_similarity(self, vec1, vec2):
	"""Calculate cosine similarity between two vectors."""
	v1, v2 = np.array(vec1), np.array(vec2)
	return np.dot(v1, v2) / (np.linalg.norm(v1) * np.linalg.norm(v2))

	def search_cache(self, embedding, threshold=0.80):
	"""
	Returns the most similar response if it is above the threshold. Otherwise returns `None`.
	"""
	k = 1

zseta / vectorize.py

Created November 5, 2025 18:50

	# pip install sentence-transformers
	from sentence_transformers import SentenceTransformer
	embedding_model = SentenceTransformer('all-MiniLM-L6-v2')
	embedding = embedding_model.encode(text).tolist()

zseta / cache_schema.cql

Created November 5, 2025 18:37

	CREATE KEYSPACE semantic_cache WITH replication = {'class': 'NetworkTopologyStrategy', 'replication_factor': '3'};

	CREATE TABLE semantic_cache.prompts (
	prompt_id uuid PRIMARY KEY,
	inserted_at timestamp,
	prompt_text text,
	prompt_embedding vector<float, 384>,
	llm_response text,
	updated_at timestamp
	);

zseta / schema.cql

Created October 7, 2025 15:55

	CREATE KEYSPACE recommend WITH replication = {'class': 'NetworkTopologyStrategy', 'replication_factor': '3'};

	CREATE TABLE recommend.movies (
	id INT,
	release_date TIMESTAMP,
	title TEXT,
	tagline TEXT,
	genre TEXT,
	imdb_id TEXT,
	poster_url TEXT,

zseta / recommender.py

Created October 7, 2025 15:51

	from db.scylladb import ScyllaClient
	from embedding_creator import EmbeddingCreator
	from models import Movie

	class MovieRecommender:

	def __init__(self):
	self.scylla_client = ScyllaClient()
	self.embedding_creator = EmbeddingCreator("all-MiniLM-L6-v2")

zseta / embedding.py

Created October 7, 2025 15:47

	from sentence_transformers import SentenceTransformer
	class EmbeddingCreator:
	def __init__(self, model_name: str = 'all-MiniLM-L6-v2'):
	self.embedding_model = SentenceTransformer(model_name, device='cpu')

	def create_embedding(self, text: str) -> list[float]:
	"""
	Get embedding for a single text input using SentenceTransformer.
	Returns the embedding vector.
	"""

zseta / scylla-ttl.cql

Created September 11, 2025 13:42

Code used for this video: https://youtu.be/SXkbu7mFHeA

	CREATE KEYSPACE ttl_example WITH replication = {'class': 'NetworkTopologyStrategy', 'replication_factor': 1};

	USE ttl_example;


	-----------------------------------------------------------
	-- BASIC USAGE
	-----------------------------------------------------------

	-- Set deafult TTL with CREATE TABLE

zseta / linux_csv.md

Created April 21, 2024 15:53

Working with large CSV files in Linux to generate database schema

Output the first N numer of rows from a csv into another:

head -n 5 data.csv > header.csv

Generate schema from csv

pip install csvkit

csvsql -d "," -q '"' -i postgresql spotifyschema.csv &gt; spotify_schema.sql

zseta / backup_docker_ssh.md

Created January 2, 2024 20:07

Backup Docker PostgreSQL database using SSH connection

Backup process with vanilla PostgreSQL or TimescaleDB

Connect to remote instance using SSH:

ssh example.com

List running Docker containers:

docker ps