[HN Gopher] 90%
       ___________________________________________________________________
        
       90%
        
       Author : bkolobara
       Score  : 53 points
       Date   : 2025-09-29 10:57 UTC (12 hours ago)
        
 (HTM) web link (lucumr.pocoo.org)
 (TXT) w3m dump (lucumr.pocoo.org)
        
       | senko wrote:
       | Since the title isn't very informative, here's a tldr:
       | 
       | > Is 90% of code going to be written by AI? I don't know. What I
       | do know is, that for me, on this project, the answer is already
       | yes. [...] At the same time, for me, AI doesn't own the code. I
       | still review every line, shape the architecture, and carry the
       | responsibility for how it runs in production. But the sheer
       | volume of what I now let an agent generate would have been
       | unthinkable even six months ago.
       | 
       | Written by Armin Ronacher of Flask, Jinja, and general Python
       | fame.
        
       | nabla9 wrote:
       | > I still review every line, shape the architecture, and carry
       | the responsibility for how it runs in production. But the sheer
       | volume of what I now let an agent generate would have been
       | unthinkable even six months ago.
       | 
       | >That said, none of this removes the need to actually be a good
       | engineer. If you let the AI take over without judgment, you'll
       | end up with brittle systems and painful surprises (data loss,
       | security holes, unscalable software). The tools are powerful, but
       | they don't absolve you of responsibility.
       | 
       | I feel the same. AI is "the code monkey". Like very inexperienced
       | that works hard and fast, has learned a lot but can't put it into
       | practice. They need constant supervision and review.
       | 
       | This will be very challenging for inexperienced programmers.
       | Normally learn by coding. You write code for fun or for money,
       | get review from more experienced, ask questions and improve. Now
       | a new programmer is expected to review AI generated code and
       | learn programming and managing AI.
        
       | vivzkestrel wrote:
       | AI cannot do anything meaningful beyond basic functions and
       | boilerplatey stuff with moderate difficulty. Don't believe me?
       | Ask your AI model to generate a production grade typescript 5.x
       | application with a tsconfig.json file that uses ts-node instead
       | of tsx with path aliases configured and uses biome 2.x for
       | linting formatting. Add vitest for testing library with graceful
       | shutdowns that handle SIGTERM and SIGINT events from your express
       | server wih an ioredis connection that can be shut down gracefully
       | along with using pg-promise to open a postgresql connection. Make
       | it use @dotenvx/dotenvx for managing development, testing,
       | staging and production environments. Add commitlint to ensure all
       | git commit messages adhere to a specific format. Ask it to add
       | lefthook along with a configuration file for running, lint format
       | and test jobs on every commit. Now add pino and pino-http with
       | custom serializers to redact passwords, tokens and apiKeys from
       | the logging output. Development version of the logs should be
       | pretty printed, production version of the logs should be
       | optimized JSON sent to a cloudwatch log transport. Add CORS to
       | handle frontends running on localhost:5173 for development,
       | https://example.com for production and add helmet for managing
       | security loopholes. Add CSRF protection to all POST, PATCH and
       | PUT routes in the application. Generate 4 different docker files,
       | one for development, one for staging, one for testing and one for
       | production. The development and testing versions should run a
       | separate postgres and redis container with whom we have seamless
       | networking setup. Development versions use self signed SSL
       | certificates at every level. The staging and production
       | environments should connect with an actual elasticache and RDS
       | instance supplied. Both staging and production use actual SSL
       | certificates. We also need Github actions for CI / CD with
       | templates for pull requests, feature requests and tasks to run
       | tests on PR with dependabot updates completely disabled. We need
       | NGINX running as forward proxy to this endpoint with SSL
       | certificates generated by letsencrypt that are auto rotated. Talk
       | to me when your AI model can generate 10% of this setup
        
         | bretpiatt wrote:
         | I'm pretty much doing that in a containerized deployment for
         | project I'm looking to open source soon called webbin:
         | Technology Stack        Frontend: React 18 + TypeScript + Vite
         | + TailwindCSS        Backend: Node.js + Express + Clustering
         | Database: PostgreSQL 16 with performance optimization
         | Cache: Redis 7 with active defragmentation        Security:
         | HTTPS/TLS with container-to-container encryption
         | Orchestration: Docker Compose with health checks
         | Monitoring: Built-in APM and performance tracking
         | Services        webbin-frontend - React TypeScript frontend
         | with HTTPS        webbin-backend - Node.js API with clustering
         | support        webbin-postgres - PostgreSQL 16 with performance
         | tuning        webbin-redis - Redis 7 with advanced caching
         | webbin-certbot - SSL Certificate management, openssl for dev,
         | LE for production        webbin-testrunner - Testbot
         | webbin-nginx - Proxy            Access Points        Frontend:
         | https://localhost:5173        Backend API:
         | https://localhost:3001        Health Check:
         | https://localhost:3001/api/health
         | 
         | I'm around 15k LOC, all built in ~80 hours of interactive
         | prompting mostly with Claude 4.0 Sonnet, then some Gemini 2.5
         | Pro for more devops activities.
        
         | turtlebits wrote:
         | I sincerely hope that isn't your prompt. These boilerplate
         | tasks are easy for an AI. You just can't expect to it to get it
         | right all in one shot. You certainly wouldn't do all that in a
         | single commit or even PR.
        
       | xigoi wrote:
       | Saying that AI is good because it produces 90% of code is like
       | saying that newsletters are good because they produce 90% of
       | e-mails.
        
       | mwkaufma wrote:
       | Hype for Bloatware.
        
       ___________________________________________________________________
       (page generated 2025-09-29 23:01 UTC)