Move Beyond Crap Code and Confusion
Frustration is easy when you're just starting out
Frustration is easy when you're just starting out
You know AI could make you more productive and help with coding, writing, and research - but there's this nagging worry about getting it wrong. What if you mess up a project because you trusted AI too much? What if you ask your AI tool something and it makes you look clueless in front of your team?
The tutorials are either too basic ("AI is amazing!") or dive straight into technical jargon about neural networks and transformers. You're stuck in this frustrating middle ground where you want to step into the future, but you also don't want to screw anything up.
The gap between what AI promises and what you can actually deliver feels intimidating, and you're not sure how to bridge it safely.
The fear of being left behind is real, but so is the fear of jumping in too fast
The fear of being left behind is real, but so is the fear of jumping in too fast
You're a solid developer. You've got skills, experience, and you deliver quality work. But there's this tension between wanting to embrace the future and being scared of messing up what's already working.
What if AI writes buggy code that breaks production? What if you become too dependent on it and lose your edge? What if you invest time learning these tools and they turn out to be just hype? The uncertainty is paralyzing, and every day you don't make a decision feels like you're either missing out on a game-changer or dodging a bullet.
I've been through this. I went from uncertainty to inspiration.
I've been through this. I went from uncertainty to inspiration.
Just over two years ago, when GitHub Copilot was in pre-release, I was asked to try out the VS Code integration and give some feedback. I hated it. It constantly interrupted me, broke my concentration, and was less than useful.
In early 2024 I began working with the VS Code team directly, helping them evolve Copilot through everyday use, trial and error, and blind exploration into "what's possible".
Eventually, I began to see patterns forming. Ways to use Copilot and other emerging AI tools for more than just code generation. Pretty soon I was seeing overall gains with my own work as well as internal engineering teams.
The key wasn't just learning the tools. It was discovering the patterns through hands-on exploration: how to structure problems, write effective prompts, review AI output, and integrate AI into development workflows.
A Deeper Look...
A Deeper Look...
Here are a few more details of what you get with both editions of the workshop, and why I created it the way I did.
A safe, confident approach to stepping into the AI future.
A safe, confident approach to stepping into the AI future.
I have wrapped up everything I've learned over the last 2 years into a single workshop that will help you onboard into the world of AI without losing your mind.
AI is a tool like any powerful development tool - with respect, proper safeguards, and clear boundaries, it can help you tremendously. For instance: if you don't want code generated for you, that's fine! Tell the LLM to create a stub with comments which help you think through the coding process.
There are so many "tricks" and patterns just like this one, and you'll learn everything that I've discovered on my journey into the AI world. Did you know you can ask your LLM to summarize your chat session into a markdown document so you have an audit record to show your boss and team what you did? You can also use code templates so you're never surprised at what the LLM will generate...
Learning through exploration
Learning through exploration
This isn't an online course, it's a self-guided workshop that works like an open world video game. I'll give you a few quest lines, you go off and explore what seems interesting.
To get you started, I've provided a 140-page PDF complete with explanations, labs, and personal insights along with over 90 minutes of video walkthroughs. In all of this, you will learn:
Simple confidence-building wins
So you can see AI's value without the fear, which means you'll actually embrace it instead of avoiding it due to uncertainty and misinformation. AI can be a powerful tool, find out why.
Context and reliability techniques
Avoiding crap code is easy if you take the time, upfront, to let your AI tool know what you expect. This can go well beyond words in a markdown file, you can use actual code, SQL, or JSON files too.
Things other than code generation
Real efficiency gains can be realized with things other than code generation, including scripting and utility functions, Git commit message generation and more. You can even troll your colleagues.
Code review that builds your confidence in the tooling
Yes, you can review the code yourself or ask a colleague to help out, but there are ways to get the tool to review itself, saving you even more time.
Putting handcuffs on the LLM with instructions and templates
The simplest, most powerful technique that people don't know about. Tell the LLM precisely what you want to see and how you want it written, and never receive crap code.
Testing strategies that help, not hinder
Build your test suite quickly using a combination of instructions and templates, but write the actual implementation yourself. LLMs are brilliant at helping you test your code confidently.
Jump in. Let's get started.
Jump in. Let's get started.
I've spent months refining this workshop, testing every technique with real developers on real projects, and building in the safeguards that let you embrace AI without the fear. This isn't about jumping on a bandwagon - it's about confidently stepping into the future of development.
The Personal Edition
You get immediate access today with a full download of all materials:
- A 140 page PDF packed with labs and explanations.
- 90+ minutes of video walkthroughs, covering each lab.
- Code for every lab, both start and finish.
- Bonus: 52 of the most common prompts that you can study and reuse, as you need.
- Bonus: The Philosophy of Prompting, an essay on structuring your thoughts for a computer, and the archetypes that follow.
If you're a solo developer or just buying for yourself, you'll be up to speed in no time, ready to help your team.
Team Edition (up to 10 Seats)
for your team, with everything you see above, plus:
- An instructor's guide on how to give the workshop to your team, what to look out for, and places you can customize.
- GitHub issue and PR templates that drive AI compliance.
- Guides for lead developers on how to handle hesitance, measure productivity, and conduct code reviews.
- A presentation on executive adoption with accompanying markdown that you can change.
- AI rollout plan templates for small, medium, and large teams.
- AI development policy templates that you can use today, for small, medium, and large teams.
- 40 style guides formatted for LLM instructions. These include popular languages, frameworks, practices and technologies that you can drop right into your project and use today.
This training goes far beyond what you'll find anywhere, giving the knowledge and the tools to get going today.
Enterprise Edition (let's talk)
This workshop can scale to organizations of any size, and is based on the internal training I gave at Microsoft to engineering teams numbering in the hundreds.
If you have a large team, I'm here to help. You'll get all of the assets above, including:
- 4 hours of my direct time, adjustable as needed, helping you ramp up your training effort and customizing the content as needed.
- Email support for a year, with a 24-hour response time.
- Virtual delivery, if needed.
- In-person delivery, if needed.
If you're interested, send me a note and I'll respond as quick as I can. I'll set up a customized package for you and your organization that will fit well within your training budget, and you'll realize efficiency gains within days.