prompt-chaining

Experimental interface environment for open source LLM, designed to democratize the use of AI. Powered by llama-cpp, llama-cpp-python and Gradio.

Multi-agent AI skill orchestration — 13 skills, 7 lifecycle phases, from idea to post-release. Works with Claude Code, Codex, Gemini CLI, Cursor, Windsurf, OpenCode, and any agent that reads Markdown or speaks MCP.

Multi-Agent LLM Framework for a research assistant for PhD scholars

10 hands-on demos covering every major agentic workflow design pattern — built with the OpenAI Agents SDK

A command-line tool that runs a series of interconnected AI prompts to automate complex writing and analysis tasks.

Cohort work for Claude Certified Architect Prep (DevCompass) — notes, labs, and solutions.

Generate a complete content package from one topic — blog post, social media captions, and email copy. Powered by a five-step prompt chaining pipeline using Groq LLM.

Safe synthetic benchmark for distributed prompt chaining in agentic AI workflows using route-first mock tool selection, propagation metrics, and reproducible ASR evaluation.

Educational Python project demonstrating a 4-stage LangChain prompt chain for recipe generation and review.

HypothesisTester.io is an interactive, AI-assisted tool built with Streamlit that enables data scientists, analysts, and students to perform hypothesis testing seamlessly. It combines statistical rigor with intuitive UI and visual explanations.

A practical portfolio and meta-analysis of advanced prompt engineering. This project not only demonstrates core techniques (ToT, CoD, Prompt Chaining) but also documents the development of a 'Master Workflow' designed to overcome inherent AI behavioral biases and limitations. Includes interactive Streamlit app with prompt playground.