Skip to content
View colesmcintosh's full-sized avatar
26 followers · 65 following

Achievements

Achievement: Pair Extraordinairex2Achievement: Pull Sharkx2Achievement: QuickdrawAchievement: YOLO

Achievements

Achievement: Pair Extraordinairex2Achievement: Pull Sharkx2Achievement: QuickdrawAchievement: YOLO

Block or report colesmcintosh

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
colesmcintosh/README.md

LLMs, Multi-Agent Systems, and AI Research

I'm Cole McIntosh, on a mission to push the boundaries of large language models and multi-agent systems. As a contributor to LiteLLM and the creator of the official LangChain Salesforce Integration, I explore innovative AI research and solutions.

Research Focus

  • Scalable, production-grade multi-agent architectures
  • High-performance LLM optimization at scale
  • Vector search and similarity optimization

Featured Projects

  • chain-of-thought-reranking: an approach to optimizing large language model (LLM) responses by extracting, reranking, and refining their internal chain-of-thought (CoT).
  • langchain-salesforce: Langchain integration for Salesforce CRM, providing tools for SOQL queries, object schema inspection, and CRUD operations through Langchain's framework.
  • vector-vault: a high-performance vector similarity search engine with LSH (Locality-Sensitive Hashing) optimization, written in Go.
  • pycuda-numpy-vector-ops: accelerating NumPy vector operations with PyCUDA in Jupyter notebooks.
  • logit-control: control LLM token generation by directly manipulating logits to enforce structured outputs, demonstrated using Hugging Face Transformers and Qwen2.5-0.5B.
  • social-media-agent-system: a sophisticated multi-agent system combining specialized AI agents to create engaging social media content, leveraging a research agent for gathering accurate, up-to-date information.

Explore these projects and star your favorites!

Pinned Loading

  1. chain-of-thought-reranking chain-of-thought-reranking Public

    an approach to optimizing large language model (LLM) responses by extracting, reranking, and refining their internal chain-of-thought (CoT). By focusing on the most coherent and relevant parts of t…

    Jupyter Notebook 4

  2. langchain-salesforce langchain-salesforce Public

    Langchain integration for Salesforce CRM, providing tools for SOQL queries, object schema inspection, and CRUD operations through Langchain's framework

    Python 3 1

  3. vector-vault vector-vault Public

    A high-performance vector similarity search engine with LSH (Locality-Sensitive Hashing) optimization, written in Go.

    Go

  4. pycuda-numpy-vector-ops pycuda-numpy-vector-ops Public

    Accelerating NumPy Vector Operations with PyCUDA

    Jupyter Notebook

  5. logit-control logit-control Public

    Control LLM token generation by directly manipulating logits to enforce structured outputs. Built with Hugging Face Transformers and demonstrated using Qwen2.5-0.5B.

    Python

  6. social-media-agent-system social-media-agent-system Public

    A sophisticated multi-agent system that combines specialized AI agents to create engaging social media content. The system leverages a Research Agent for gathering accurate, up-to-date information …

    Python