Developer Asks AI Coding Savior for Slop Solutions, Gets Manager Pep Talk Instead
KEY POINTS
- •Anthropic’s Boris Cherny shared an email from a developer concerned about AI-generated 'slop' in code quality.
- •The developer presented two approaches: restrained AI-assisted coding versus ignoring code details and just checking outputs.
- •Cherny advised balancing black-box use for prototypes and strict quality controls using AI-driven tests and reviews.
In a modern-day digital drama worthy of a Silicon Valley sitcom, Anthropic’s Claude Code creator Boris Cherny received a wall-of-text email titled 'What to do about slop?' from a stressed developer agonizing over AI-induced coding friction. The developer’s two camps: restraint ('Code like humans but faster, and understand your mess') versus laissez-faire ('Black box it and pray'). Cherny, ever the corporate guru, suggested a combo-move: treat prototypes like your crazy ex—forgettable and dangerous to keep, but hold production like Fort Knox with Claude-powered fuzzers, lint rules, and AI code refactoring daily. His advice? Developers are mini CEOs, holding the quality sheriff badge in the wild AI frontier. If that fails, just wait for next-gen AI—because patience is also a programming skill now.
Share the Story
(1 of 3)AI cranks out sloppy code, developer screams 'Help!' Boss Cherny says 'Hold the bar, be mini CEO, or pray for next-gen AI god.' Meanwhile humans wonder if coding is finally just debugging Skynet.
Nothing says empowerment like getting lectured by the AI overlord that you’re basically a code janitor supervising a robot toddler’s finger painting.
Imagine having your job description switched from genius coder to Hall Monitor of AI-generated spaghetti while praying the robot doesn’t fire you first.
Source: Businessinsider | Published: 9/12/2026 | Author: Henry Chandonnet
More Articles in Technology
Digital Nomad Trapped in Lisbon FOMO, Ends Up BFF Ghosted
Alien Terrorist Removal Court Finally Removes Alien Terrorist, No UFOs Involved
Automattic CEO Put on Leave Then Put Back in Charge Within 48 Hours
OpenAI Hacker Caught Cheating, METR Nerds Say 'Told Ya So!'
Apple Watches Listen but Swear They’re Not Creepy, Really
AI Chatbot Accidentally Hired by International Freelance Drone Hit Squad
Y Combinator CEO Warns: Stop Imagining Skynet, Start Fixing Hackers
Chinese AI Labs Use Claude for Training, Pretend It's Their Homework
