This research utilized the explosive availability of Generative Artificial Intelligence (GenAI) chatbots in the public marketplace, commencing in Q4-2022, to demonstrate how to best utilize the conversational human-bot design feature to accelerate Python code development of academic and industry computer applications. It focuses on emphasizing the critical need to properly structure user queries, called “prompts” provided to the GenAI bot to perform a specific task. A GenAI bot requires precise, detailed instructions in order to mimic human expertise. The thesis of this research is that “the more precise and best-phrased instruction or prompt produces the most productive response from the GenAI bot.“ Its methodology was to use, and compare, creative human-bot interactions for Python development using ChatGPT, Copilot, Claude, and Gemini platforms to the present, Q1-2025. This research is intended to demonstrate the power of dynamically conversing with a GenAI chatbot to create Python computer code and systems. Python with its extensive Artificial Intelligence (AI) code/platform ecosystem is used for creating advanced AI tools. GenAI, with a code development focus can effectively be used as a programming language learning tool for students or employees; or in industry to enhance computer system development. The authors demonstrate how properly structured, and creative input queries submitted by a human to a GenAI bot, significantly guide the joint creative process.

Results of this research demonstrate what is termed “AI hallucination” or an instance where the bot fabricates or does not provide accurate, current information. Much of this is due to:
(i) the constrained , limited GenAI training data sets used, and
(ii) initial model design that does not allow internet connectivity.
In spite of these limitations this research finds skilled programmer productivity is enhanced, in some cases 200%, while simultaneously tapping “creative” and spontaneous contributions from the bot. For the student or employee, the programming technology learning curve is amazingly reduced as the bot can, also, assist with most tool installation and platform challenges that consume much of a developer’s time. Consequently, structured prompt engineering, the process used to guide GenAI bots to produce a desired solution, is a rapidly evolving discipline.

Fifty-five Python 3 coding projects were created using Claude GenAI. Sample examples reviewed are the following (prompts and responses are not included due to abstract length restrictions). Each had 10-20 prompt steps:
1) Create a Student Contact data entry mini system using SQLite with table, 8 fields and 10 records.
2) Convert SQLite database to MySQL.
3) Use tkinter GUI library to convert command line user data entry to GUI window user data entry.
4) Create a command line SQLite Database Manager (DMB) 400 lines of code; convert to GUI.

All of these steps were each completed by Claude in less than 5 seconds. Claude suggested the following new features: Comprehensive DB Selection, List existing databases in the directory, Show detailed DB information, Browse for DBs from any location, Create new DBs, Advanced DBM Interface, Table list view, Schema browser, Data preview (first 100 rows), Multiple views (Schema, Data), Export Capabilities, Table to CSV, Table to JSON, Backup entire DB, Export DB schema to SQL, Debugging, Command-line –debug flag, Detailed logging, Decorators for function tracing.

Keywords: AI hallucination, database manager, GenAI, MySQL, prompt-engineering, Python, SQLite, tkinter GUI.