I’m a 20-Year Engineer – AI Coding Tools Are My New Oxygen (But They’re Toxic If You Breathe Too Deep)
Let’s be clear: I'm totally hooked on coding with AI and not looking back. Here’s how to use it like a stimulant, not a crutch AI coding assistants aren’t just helpful – they’re a cognitive exoskeleton. I code 3x faster with Cursor, ship prototypes in hours instead of days, and offload the work that makes my brain feel like overcooked ramen. But here’s the catch: every line of AI code is a Faustian bargain. Why I’m Addicted (And You Should Be Too) AI isn’t “helpful” – it’s a force multiplier for: Killing yak shaves (Bash scripts, config hell, regex, TypeScript puzzlers, SQL gymnastics) Bypassing syntax paralysis (How do you format dates in Swift again? Why wouldn't this API cooperate?) Prototyping at methamphetamine speed (Need a React table with client-side sorting? Generated in 8 seconds) Last month, I prototyped a RAG tool for my AI agent with 87% AI-generated code (local vector DB, embeddings, ranking – the whole circus). 2 hours instead of 2 days. Was it perfect? Hell no. But it let me start iterating immediately. Now that we're shipping? I’m rewriting 87% manually – with AI as my WD-40 for stubborn code bolts. The Tech Debt Paradox AI-generated code isn’t “bad” – it’s strategically irresponsible. I treat it like a high-interest credit card: SWIPE FREELY for: Throwaway code (one-off scripts, spike solutions) Boilerplate (mocks, DTOs, CRUD skeletons) “I just need this to work once” moments (local DB migrations) PAY DOWN IMMEDIATELY when: It touches business logic Performance/security/privacy matter (read: production) You’re past exploration and need precision Example: Last month Claude wrote 5 Jest mocks + 10 unit tests for an unfamiliar codebase. 45 minutes saved. 2 hours of existential crisis avoided. Tech debt? 15 minutes fixing incoherent tests. Net win!
Let’s be clear: I'm totally hooked on coding with AI and not looking back. Here’s how to use it like a stimulant, not a crutch
AI coding assistants aren’t just helpful – they’re a cognitive exoskeleton. I code 3x faster with Cursor, ship prototypes in hours instead of days, and offload the work that makes my brain feel like overcooked ramen. But here’s the catch: every line of AI code is a Faustian bargain.
Why I’m Addicted (And You Should Be Too)
AI isn’t “helpful” – it’s a force multiplier for:
Killing yak shaves (Bash scripts, config hell, regex, TypeScript puzzlers, SQL gymnastics)
Bypassing syntax paralysis (How do you format dates in Swift again? Why wouldn't this API cooperate?)
Prototyping at methamphetamine speed (Need a React table with client-side sorting? Generated in 8 seconds)
Last month, I prototyped a RAG tool for my AI agent with 87% AI-generated code (local vector DB, embeddings, ranking – the whole circus). 2 hours instead of 2 days. Was it perfect? Hell no. But it let me start iterating immediately. Now that we're shipping? I’m rewriting 87% manually – with AI as my WD-40 for stubborn code bolts.
The Tech Debt Paradox
AI-generated code isn’t “bad” – it’s strategically irresponsible.
I treat it like a high-interest credit card:
-
SWIPE FREELY for:
- Throwaway code (one-off scripts, spike solutions)
- Boilerplate (mocks, DTOs, CRUD skeletons)
- “I just need this to work once” moments (local DB migrations)
-
PAY DOWN IMMEDIATELY when:
- It touches business logic
- Performance/security/privacy matter (read: production)
- You’re past exploration and need precision
Example: Last month Claude wrote 5 Jest mocks + 10 unit tests for an unfamiliar codebase. 45 minutes saved. 2 hours of existential crisis avoided. Tech debt? 15 minutes fixing incoherent tests. Net win!
