Specializing in LLMs and AI, I build end-to-end applications spanning modern front-end interfaces and scalable back-end systems. I architect AI-powered solutions including agentic workflows, RAG pipelines, and LLM-based data extraction frameworks. I work across the full technology stack, deploying production systems on AWS, Azure, and GCP, and contribute to open-source AI frameworks that reduce complexity through clear abstractions and intentional design.
Projects
An open source extraction framework that is LLM agnostic and uses agentic search to parse data from long documents.
An applied guide to steering LLM generations with logit bias techniques and reusable tooling.
A technique to optimize LLM responses by extracting and refining their internal reasoning chains.
How to query an AWS Bedrock Knowledge Base using the RetrieveAndGenerate API.
A LangChain integration package that enables seamless interaction with Salesforce CRM data in LLM applications.
A Model Context Protocol server for performing numerical computations through LLMs using NumPy.
Blog
View all →Why Fast, Small Models Are the Way to Go
LLMs fail. The question is not whether your model will fail, but how gracefully and how quickly. Fast failures beat slow failures every time.
Why Open-Weight Models Matter for AI Independence
Why owning open-weight models protects builders from account bans, rate limits, and centralized policy shifts that can kneecap innovation overnight.
The Anti-Slop Backlash: Keep Thinking in a Synthetic Sludge Age
A call to resist automated brain rot, expose the hollow economics of generative visuals, and double down on language-first tools that sharpen judgment.
Open Source
Popular framework for building LLM-powered applications and agents.
Block's Goose project, an open source initiative for AI Agents.
Huggingface's project focused on lightweight, efficient AI agents.
Open-source LLM Gateway that allows calling 100+ LLM APIs in the OpenAI format.
Work
Resume
Connect
Feel free to contact me at cole@staymellow.ai