> Markdown version of [/videos/631-chatgpt-create-a-presentation?t=785](https://www.wearedevelopers.com/videos/631-chatgpt-create-a-presentation?t=785). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# ChatGPT: Create a Presentation!

Your engineering job is safe because clients cannot exactly describe what they want. Learn how to safely leverage ChatGPT for test-driven development, complex refactoring, and SQL queries.

- **Speakers:** [Markus Walker](https://www.wearedevelopers.com/@markus-walker)
- **Event:** World Congress 2023
- **Published:** August 11, 2023
- **Duration:** 30:24
- **URL:** https://www.wearedevelopers.com/videos/631-chatgpt-create-a-presentation

## Summary

This session explores the practical applications of ChatGPT for software development, shedding light on its strengths, limitations, and peculiar behaviors. Operating as an auto-regressive language model rather than a computational engine, ChatGPT excels at pattern recognition but often struggles with hard math or algorithmic logic. Through live demonstrations, the presentation highlights how generative tools can rapidly analyze obfuscated scripts, refactor concepts like Conway's Game of Life, translate generic code to TypeScript, and deconstruct complex regular expressions. However, it also exposes the technology's unreliability with algorithmic minutiae, demonstrating how the model confidently misjudges memory complexity or fails numerical logic puzzles.

Developers can leverage these tools creatively for test-driven development by feeding the AI failing unit tests to generate satisfying functions, or even utilize the platform as an interactive mock SQL Server for sandbox database queries. Advanced use cases involve treating ChatGPT as a prompt engineer to refine its own contextual instructions, alongside exploring infrastructure integrations like generating Terraform templates via natural language.

Crucially, this technology functions as an interactive repository of community code, carrying the exact same inherent risks of unchecked copy-pasting. Engineers must employ highly specific instructions, explicitly define negative constraints, and meticulously validate all outputs to avoid integrating snippets that fail to build or introduce severe security vulnerabilities. Ultimately, despite the rapid advancement of automated tooling, developer roles remain firmly secure for a fundamentally human reason: clients must be able to "exactly describe what they want, so we are safe."

**Keywords:** chatgpt software development, large language models, auto-regressive models, javascript refactoring tools, typescript code conversion, regular expression synthesis, test-driven development, algorithm memory complexity, sql server database mocking, prompt engineering constraints, infrastructure as code tools, terraform ai integration, github copilot limits, codebase security vulnerabilities, developer productivity workflows

## Chapters

1. **Introduction to AI history and developer chat tools** (02:01) — How advancements from image recognition set the stage for modern text generators.
1. **Understanding capabilities and limitations of large language models** (04:09) — How the auto-regressive text generator predicts sentences but lacks actual computational math logic.
1. **Analyzing and identifying complex code structures with AI** (07:02) — How to pre-prompt the assistant to identify obfuscated code like Conway's Game of Life.
1. **Refactoring and porting legacy code to new languages** (08:04) — Translating vanilla JavaScript implementations into well-documented TypeScript applications.
1. **Deconstructing and explaining complex regular expressions with AI** (09:27) — Breaking down complex regular expressions to understand capturing groups and matching rules.
1. **Generating and testing regular expressions with technical prompts** (10:32) — Creating robust regex patterns for primes and phone numbers despite initial generation flaws.
1. **Evaluating generated sorting algorithms and application memory complexity** (13:05) — Assessing the accuracy of generated quicksort algorithms and uncovering hallucinated performance metrics.
1. **Navigating complex math puzzles and AI logic hallucinations** (16:06) — How the model struggles with mathematical constraints and fabricates incorrect console outputs.
1. **Implementing algorithms through test-driven development methodologies and prompts** (18:45) — Feeding existing unit tests into the model to generate satisfying function implementations.
1. **Refining complex software instructions using automated prompt engineers** (21:19) — Using the model to refine initial phrasing and generate structured prompts for better results.
1. **Simulating interactive virtual database environments for query testing** (23:05) — Prompting the model to act as a relational database to execute tables and mock data queries.
1. **Exploring AI integrations in modern agile development workflows** (26:42) — Highlighting popular tools that embed language models directly into standard environments.
1. **Key takeaways and security risks of generated code** (28:30) — Guidelines on validating outputs to prevent security vulnerabilities and address job replacement realities.

## Related Moments

- [Running a live messaging demonstration in ChatGPT](https://www.wearedevelopers.com/videos/1498-build-your-first-ai-assistant-in-30-minutes-no-code-workshop) (from "Build Your First AI Assistant in 30 Minutes: No Code Workshop")
- [Launching a ChatGPT driver's license for HR professionals](https://www.wearedevelopers.com/videos/1356-from-learning-to-leading-why-hr-needs-a-chatgpt-license) (from "From Learning to Leading: Why HR Needs a ChatGPT License")
- [Introduction to AI code generation and developer habits](https://www.wearedevelopers.com/videos/600-stack-overflow-community-and-ai) (from "Stack Overflow: Community and AI")
- [Generative artificial intelligence and programming fundamentals](https://www.wearedevelopers.com/videos/1291-using-all-the-html-running-state-of-the-browser-and-modern-is-rubbish) (from "Using all the HTML, Running State of the Browser and "Modern" is Rubbish")
- [Audience Q&A on tooling choices and AI application prototyping](https://www.wearedevelopers.com/videos/100195-xcode-development-redefained) (from "Xcode development redefAIned")
- [Exploring popular generative AI models and applications](https://www.wearedevelopers.com/videos/844-enter-the-brave-new-world-of-genai-with-vector-search) (from "Enter the Brave New World of GenAI with Vector Search")

## Related Articles

- [WWC24 Talk - Scott Hanselman - AI: Superhero or Supervillain?](https://www.wearedevelopers.com/magazine/469-wwc24-talk-scott-hanselman-ai-superhero-or-supervillain)
- [MLOps And AI Driven Development](https://www.wearedevelopers.com/magazine/82-mlops-and-ai-driven-development)
- [The Top ChatGPT Plugins for Developers in 2023](https://www.wearedevelopers.com/magazine/259-the-top-chatgpt-plugins-for-developers-in-2023)
- [Dev Digest 108 - Git off my cloud!](https://www.wearedevelopers.com/magazine/407-dev-digest-108-git-off-my-cloud)

## Related Jobs

- [AI Software Engineer (Germany)](https://www.wearedevelopers.com/jobs/48317-ai-software-engineer-germany) at **Sunhat**
- [Staff Developer Advocate, GitHub Security Lab](https://www.wearedevelopers.com/jobs/ext/1921051-staff-developer-advocate-github-security-lab) at **GitHub**
- [Staff Software Engineer, Copilot Experiences](https://www.wearedevelopers.com/jobs/ext/164361-staff-software-engineer-copilot-experiences) at **GitHub**
- [Principal Software Engineer, Enterprise AI Platform](https://www.wearedevelopers.com/jobs/ext/1467292-principal-software-engineer-enterprise-ai-platform) at **GitHub**
- [Principal Product Manager, Agent Platform](https://www.wearedevelopers.com/jobs/ext/277541-principal-product-manager-agent-platform) at **GitHub**
- [Senior Software Engineer](https://www.wearedevelopers.com/jobs/ext/15942-senior-software-engineer) at **GitHub**