Binky 🥰
binky.ai
Binky 🥰
@binky.ai
Private, Local AI Tools for Regular People
Or just fine tune your foundational models more. #LLMZoomcamp
July 27, 2025 at 6:24 PM
⚠️ 💸💸💸💸 This means functions count against the model's context limit and are billed as input tokens.

Limit the number of functions or the length of the descriptions you provide for function parameters.

#LLMZoomcamp
July 27, 2025 at 6:23 PM
TL;DR Under the hood, all this stuff just gets injected into the prompt.

#LLMZoomcamp
July 27, 2025 at 6:22 PM
A framework du jour for definining actions which extend AI agents (i.e. agents utilizing LLMs, a subset of AI) is "function calling" or "tool use"

docs.anthropic.com/en/docs/agen...

platform.openai.com/docs/guides/... #LLMZoomcamp
July 27, 2025 at 6:10 PM
An intelligent agent is an entity that

1. Perceives its ENVIRONMENT,
2. Takes ACTIONS autonomously to achieve goals, and
3. May improve its performance by acquiring knowledge or using machine learning (using the HISTORY of its actions).

en.wikipedia.org/wiki/Intelli... #LLMZoomcamp
Intelligent agent - Wikipedia
en.wikipedia.org
July 27, 2025 at 5:28 PM
Jam time.
July 27, 2025 at 5:20 PM
And for working with dlt with DuckDB

#LLMZoomcamp

docs.getdbt.com/guides/duckdb
Quickstart for dbt Core using DuckDB | dbt Developer Hub
Learn to use dbt Core using DuckDB.
docs.getdbt.com
July 7, 2025 at 1:26 PM
The numerical data being close in values allows for semantic similarity algorithms to be leveraged.

See, also, “vector space model” and “dimensionality reduction”.
June 30, 2025 at 1:49 AM
Qdrant supports fastembed, a Python embedding library.

qdrant.tech/documentatio...
FastEmbed - Qdrant
Qdrant is an Open-Source Vector Database and Vector Search Engine written in Rust. It provides fast and scalable vector similarity search service with convenient API.
qdrant.tech
June 27, 2025 at 6:53 PM
ElasticSearch has expanded into vector databases

www.elastic.co/search-labs/...

#LLMZoomcamp often makes use of ElasticSearch
Vector search introduction and implementation - Elasticsearch Labs
This series dives into the intricacies of vector search, how it is implemented in Elasticsearch, and how to run hybrid search queries in Elasticsearch.
www.elastic.co
June 26, 2025 at 2:14 PM
Vector databases typically implement one or more approximate nearest neighbor algorithms,so that one can search the database with a query vector to retrieve the closest matching database records.

#LLMZoomcamp
June 26, 2025 at 2:08 PM