Developing AI-enabled research and operating workflows

Context

Alongside my formal roles, I have continued building practical experiments with Python, LLMs, coding assistants, and agents.

What I did

This has included:

• developing a Python-based news aggregation pipeline using scraping, MongoDB, keyword analysis, and sentiment analysis;
• experimenting with agents that monitor logistics news and turn developments into structured analysis;
• using Codex to develop tools through written requirements, specifications, and iterative testing;
• converting operational knowledge into reusable analysis logic and decision frameworks;
• exploring how AI-generated analysis can be checked, interpreted, and translated into strategy and execution.

These projects are part of a wider shift in how I work. Rather than using AI only to produce isolated outputs, I am learning how to design the context, specifications, review mechanisms, and operating processes around it.

What this demonstrates

A practical transition from using AI as a productivity tool toward designing AI-enabled ways of working.