In the dynamic world of consulting, staying ahead means embracing innovation at every turn. My venture into artificial intelligence (AI) has been transformative. This will be my first (of many) posts related to AI. Let me guide you through my cerebral expansion as I’ve integrated AI into the various facets of my work.

I have used OpenAI in the cases cited here. Choose one that makes sense to you. It’s just what works for me.

Use ChatGPT like a ChatBot

My day-to-day work relies heavily on AI, with ChatGPT as an indispensable tool. Here are some examples of how I leverage it to enhance efficiency and accuracy:

SQL

  • Writing and Adapting SQL Queries: AI allows me to write SQL queries with precision.
  • Merging Queries: AI helps me merge SQL queries maintaining data integrity.
  • Fixing Queries: AI assists in fixing queries
  • Converting SQL Queries: Each database has its quirks. AI aids in converting SQL queries to be compatible with different databases, whether it’s MySQL, PostgreSQL, or Oracle.

Writing and Documenting

  • Documentation: Creating comprehensive documentation is made easier with AI. I provide the technical content, and AI structures it into user-friendly documents.
  • Fixing Documentation: AI helps refine existing documentation, improving grammar and intonation, ensuring clarity, and maintaining the professional tone required for technical writing.
  • Email Correspondence: AI enhances communication by drafting emails with the appropriate intonation, reflective of the context—whether it’s assertive for leadership or informative for clients.
  • Code Documentation: With just the code at hand, AI generates detailed documentation, describing functions and methodologies clearly.

Code

  • Translating Code: AI assists in translating code from one programming language to another, maintaining logical integrity.
  • Adapting Code: When existing code needs modification to meet new requirements, AI helps in adapting the code efficiently, saving valuable time.
  • Fixing Code: AI comes in handy for debugging, sometimes just copying and pasting the error it will fix for you.
  • Completing Code: AI can even suggest and implement missing methods or pieces in your code, contributing to the overall robustness of the application.
  • Writing Regex: Crafting regular expressions is a breeze with AI, as it generates precise patterns for data validation.
  • Enhancing Logs: AI improves logging in existing code, ensuring that the logs are detailed and useful for troubleshooting.
  • Refactoring Code: When it’s time to clean up a code, AI provides insights on refactoring, helping to streamline and optimize for better efficiency and readability.

These are just a few examples above and far from all that I use in my day-to-day.

This stage of AI application is about enhancing productivity in the routine yet critical aspects of my consultancy work.

Example: Creating queries without worrying about missing a parenthesis, or any other type of silly error.

Use ChatGPT to be your Assistant

As my virtual assistant, AI goes beyond mere chat interactions. It plays a crucial role.

The possibilities are endless, the practice of doing this on a daily basis leads me to always wonder if artificial intelligence could solve this more quickly so that I can move on to other tasks and solve other things.

Example 1: This is AI helping me analyze Excel sheets to detect duplicate projects/space keys

Example 2: Extract issue keys from extensive texts

Example 3: Your best ally in note-taking and meeting analysis from transcripts

Are you starting to grasp the idea? All the above prompts are real examples from real customers.

Could I do it without using AI? I could! But it would be much slower, with the risk of missing a parenthesis, missing a key, anything! It’s about using technology to your advantage, as an ally, you just have to do the thinking, let it handle the grunt work.

Create GPTs to make you even Faster and Smarter

Here’s where efficiency leaps forward. I’ve tailored GPTs to:

General Translations: Being Brazilian, these custom GPTs adeptly translate content to English, perfecting tone or even drafting tickets for Atlassian support when needed. Don’t tell me you don’t find this boring, I know it is, and using AI to help me explain makes it faster, clearer, and helps them help me too, it’s a win-win.

Coding Assistant: I created a GPT explaining everything I would like to use within the code, such as logging library, code comments, how I want authentication to be done, the entire Jira API (since it simply understands the Postman collection I put inside the GPT for it to use), and many other details. It’s incredible how quickly solutions can be created today. Soon, this precision will be so great that a person won’t need to know how to program, as long as they understand the code and know how to fit the pieces together like assembling a Lego.

Example: Build scripts a lot faster

Note that the prompt is just that, underneath are all examples of code, documents, APIs, Postman Collections, instructions, everything.

I have some other specific GPTs and still many that I will end up creating, but I am sure that if you don’t have one yet and discover how this can help you, you will understand the wonder that this is and the efficiency it brings to your work and your deliveries.

Build Solutions using OpenAI API

This is where the magic happens, I have been creating more complex solutions using the OpenAI API and learning to better connect with prompts. It’s all a learning curve.

But as an example, I created an internal solution for the company that translates entire spaces from any language to any language, taking all macros, links, attachments. Very useful for training translations. It does everything within minutes.

I already have some ideas for plugins that I would love to put here but who knows, maybe they will turn into products in the marketplace later on? We’ll see in the next chapters.

As an efficiency addict, the potential for transformation and efficiency that this will generate in the coming years is incredible, as it already has.

Train and Use my Own LLM

This is where I have been exploring the most, I still don’t think it’s close to what OpenAI is delivering in terms of ease and speed, but definitely if you have something specific and are reluctant to send your information to OpenAI, this is the only way, and it has become very easy to do with so many good tools in the market already.

Search for (LLM Studio or Ollama) + AnythingLLM and you will be surprised how easy is to create something, and it’s just starting.

Here is a basic example and prompt performed for the model TheBloke/WizardCoder with 33 billion parameters on my local machine, without any internet connection.

This is how good this model is as of March 2024:

I must say that the fact we have this at our disposal is almost bizarre.

Conclusion: Embrace AI or Be Surpassed

Adopting AI is not just about keeping pace, it’s about setting the pace. Those who don’t leverage AI risk being overtaken by those who do, and this applies to both individuals and businesses. It’s about continuous learning, humility in the face of rapidly evolving technology, and the professional drive to deliver the utmost value to clients.

As I heard on a YouTube channel and found extremely valid:

The Prompt is the new fundamental unit of building things.

I hope that with this post you have managed to understand where Artificial Intelligence will help you in your daily life, obviously, this does not need to be only focused on the area of Atlassian consulting, which is my case.