Developer Articles | TechForDev

Latest AI / ML JavaScript Python React Next.js Web Dev DevOps Cloud

Docker for Node.js Developers: From Zero to Production Without Losing Your Mind

Juan Torchia3d ago • 6 min read

Docker for Node.js Developers: From Zero to Production Without Losing Your Mind

Three broken Dockerfiles, two production outages, and one sleepless night — that's what it cost me t...

#english#tutorials#tutorial#typescript

2 0

m2cgen: export your ML model without shipping Python to production

Juan Torchia1d ago • 5 min read

m2cgen: export your ML model without shipping Python to production

Train in Python, deploy in Java, Go, or whatever you've got. m2cgen converts your scikit-learn model...

#english#experiments#machinelearning#opensource

2 0

TigerFS: A Full Filesystem Inside PostgreSQL (And Why This Obsession Feels Like a Symptom)

Juan Torchia3d ago • 8 min read

TigerFS: A Full Filesystem Inside PostgreSQL (And Why This Obsession Feels Like a Symptom)

Someone built a complete filesystem inside PostgreSQL. Last year I shoved Linux's entire git history...

#english#experiments#experimentos#postgres

0 0

Claude Design and What It Reveals About How Anthropic Thinks (or Doesn't Think) About Developers

Juan Torchia1d ago • 8 min read

Claude Design and What It Reveals About How Anthropic Thinks (or Doesn't Think) About Developers

A viral HN post with 1050 points on Claude's design made me confront something I've been feeling for...

#english#opinion#claudecode#anthropic

1 0

I Measured How Much Each Agent Design Decision Costs in Tokens (The Numbers Make Me Uncomfortable)

Juan Torchia1d ago • 9 min read

I Measured How Much Each Agent Design Decision Costs in Tokens (The Numbers Make Me Uncomfortable)

Long vs short prompts, tool calls vs plain text, accumulated vs summarized context. I measured the r...

#english#experiments#typescript#optimizacion

1 0

Do AI Agent Costs Grow Exponentially? I Ran My Logs and the Answer Surprised Me

Juan Torchia1d ago • 7 min read

Do AI Agent Costs Grow Exponentially? I Ran My Logs and the Answer Surprised Me

A Hacker News thread with 208 points asks whether AI agent costs grow exponentially. I have months o...

#english#reflections#typescript#optimizacion

1 0

A Dancer with ALS Controlled a Performance with Her Brainwaves — and I Couldn't Stop Thinking About It

Juan Torchia3d ago • 8 min read

A Dancer with ALS Controlled a Performance with Her Brainwaves — and I Couldn't Stop Thinking About It

An artist with ALS used a BCI to control a live dance performance in real time. Not rehab. Not medic...

#english#reflections#accesibilidad#opensource

0 0

Apple as the AI 'Loser' That Ends Up Winning: I Lived It When Anthropic Ghosted Me for a Month

Juan Torchia3d ago • 8 min read

Apple as the AI 'Loser' That Ends Up Winning: I Lived It When Anthropic Ghosted Me for a Month

Apple's 'accidental moat': privacy wasn't a feature — it was the Plan B nobody took seriously. I hav...

#english#reflections#ia#modeloslocales

0 0

I Reallocated $100/mo From Claude Code to Zed + OpenRouter: What Nobody Tells You About Switching AI Tools

Juan Torchia3d ago • 8 min read

I Reallocated $100/mo From Claude Code to Zed + OpenRouter: What Nobody Tells You About Switching AI Tools

This isn't just about money. When you switch AI tools, you switch your workflow. When you switch you...

#english#opinion#typescript#claudecode

0 0

Google Gemma 4 Runs Natively on iPhone: I Tested It and the Gap Between 'Works' and 'Useful' Is Still Massive

Juan Torchia3d ago • 8 min read

Google Gemma 4 Runs Natively on iPhone: I Tested It and the Gap Between 'Works' and 'Useful' Is Still Massive

Gemma 4 runs offline on iPhone. I replicated it on my 13 with a nearly full 128GB. It works. But the...

#english#experiments#inferencialocal#google

0 0

Research-Driven Agents: Making the Agent Read Before It Codes

Juan Torchia3d ago • 8 min read

Research-Driven Agents: Making the Agent Read Before It Codes

Months watching agents dump code without context and break everything. I ran a real experiment: forc...

#english#experiments#typescript#workflow

0 0

Project Glasswing: What AI Doesn't Tell You When It Writes Your Code

Juan Torchia3d ago • 8 min read

Project Glasswing: What AI Doesn't Tell You When It Writes Your Code

Glasswing hit a nerve I've had for a while. We deploy with AI, generate code with AI, and the attack...

#english#technology#devops#seguridad

0 0

Claude Code Routines: Weeks Ignoring It and I Finally Get Why It Matters

Juan Torchia3d ago • 7 min read

Claude Code Routines: Weeks Ignoring It and I Finally Get Why It Matters

I used Claude Code for weeks without setting up a single routine. Assumed it was overhead for people...

#english#experiments#typescript#claudecode

1 0

Vibe-Coding vs Stress-Coding: How I Actually Use AI on Projects That Matter

Juan Torchia3d ago • 8 min read

Vibe-Coding vs Stress-Coding: How I Actually Use AI on Projects That Matter

Vibe-coding is fantastic — until your project has real users. Here's the concrete difference between...

#english#opinion#typescript#nextjs

0 1

Bondi Sonoro: A Build Log of Real Data, Generative Music, and the MTA.me Mechanic

Juan Torchia3d ago • 13 min read

Bondi Sonoro: A Build Log of Real Data, Generative Music, and the MTA.me Mechanic

From static train GTFS to real-time bus positions. A full walkthrough of the decisions, the bugs, th...

#english#experiments

0 0

Docker Pull Fails in Spain Because of Cloudflare and a Soccer Match — Nobody Talks About the Real Pattern

Juan Torchia3d ago • 9 min read

Docker Pull Fails in Spain Because of Cloudflare and a Soccer Match — Nobody Talks About the Real Pattern

A soccer match triggered ISP blocks on Cloudflare ranges in Spain and broke Docker Hub for thousands...

#english#opinion#docker#devops

0 0

I Reviewed 3 Vibe-Coded PRs With Hardcoded Keys — The Problem Isn't the AI, It's That I Approved Them

Juan Torchia3d ago • 8 min read

I Reviewed 3 Vibe-Coded PRs With Hardcoded Keys — The Problem Isn't the AI, It's That I Approved Them

Three AI-generated PRs. Three AWS API keys sitting in the code. Three times I approved them because ...

#english#reflections#typescript#devops

0 0

How They Broke the Top AI Agent Benchmarks — and What That Says About My Stack

Juan Torchia3d ago • 7 min read

How They Broke the Top AI Agent Benchmarks — and What That Says About My Stack

I read the paper that exploded on HN about how top AI agent benchmarks get shattered. The problem is...

#english#opinion#typescript#llm

0 0

A 'perfectible' language: why the idea is beautiful and why it'll fail anyway

Juan Torchia3d ago • 9 min read

A 'perfectible' language: why the idea is beautiful and why it'll fail anyway

I read a proposal for a language designed to evolve its own syntax and couldn't stop thinking about ...

#english#opinion#typescript#programacion

0 0

pnpm vs npm vs yarn vs bun: The Real Comparison Nobody Gives You in 2025

Juan Torchia3d ago • 6 min read

pnpm vs npm vs yarn vs bun: The Real Comparison Nobody Gives You in 2025

I used all four in real projects. One wrecked a monorepo at 3am. Another saved my ass in production....

#english#technology#pnpm#npm

0 0

France Ditches Windows for Linux: What We're All Missing

Juan Torchia3d ago • 8 min read

France Ditches Windows for Linux: What We're All Missing

France announced the largest Linux migration in Europe. I spent six months with a provincial agency ...

#english#opinion#devops#linux

0 0

Security as Proof of Work: Why Compliance Saves Nobody

Juan Torchia3d ago • 7 min read

Security as Proof of Work: Why Compliance Saves Nobody

I spent weeks setting up SPF, DKIM, DMARC, Dependabot, and Snyk. And someone still approved a PR wit...

#english#reflections#devops#seguridad

0 0

Your Accessibility Score Is Lying to Your Face

Juan Torchia3d ago • 9 min read

Your Accessibility Score Is Lying to Your Face

I scored 98/100 on Lighthouse and axe on juanchi.dev. Then I asked someone who actually uses a scree...

#english#experiments#react#frontend

0 0

Multi-Agent Software Development Is a Distributed Systems Problem

Juan Torchia3d ago • 8 min read

Multi-Agent Software Development Is a Distributed Systems Problem

I read the multi-agent development paper and the penny dropped hard: the weird bugs I kept seeing in...

#english#opinion#vibecoding#multiagente

0 0

Tech Articles

Docker for Node.js Developers: From Zero to Production Without Losing Your Mind

m2cgen: export your ML model without shipping Python to production

TigerFS: A Full Filesystem Inside PostgreSQL (And Why This Obsession Feels Like a Symptom)

Claude Design and What It Reveals About How Anthropic Thinks (or Doesn't Think) About Developers

I Measured How Much Each Agent Design Decision Costs in Tokens (The Numbers Make Me Uncomfortable)

Do AI Agent Costs Grow Exponentially? I Ran My Logs and the Answer Surprised Me

A Dancer with ALS Controlled a Performance with Her Brainwaves — and I Couldn't Stop Thinking About It

Apple as the AI 'Loser' That Ends Up Winning: I Lived It When Anthropic Ghosted Me for a Month

I Reallocated $100/mo From Claude Code to Zed + OpenRouter: What Nobody Tells You About Switching AI Tools

Google Gemma 4 Runs Natively on iPhone: I Tested It and the Gap Between 'Works' and 'Useful' Is Still Massive

Research-Driven Agents: Making the Agent Read Before It Codes

Project Glasswing: What AI Doesn't Tell You When It Writes Your Code

Claude Code Routines: Weeks Ignoring It and I Finally Get Why It Matters

Vibe-Coding vs Stress-Coding: How I Actually Use AI on Projects That Matter

Bondi Sonoro: A Build Log of Real Data, Generative Music, and the MTA.me Mechanic

Docker Pull Fails in Spain Because of Cloudflare and a Soccer Match — Nobody Talks About the Real Pattern

I Reviewed 3 Vibe-Coded PRs With Hardcoded Keys — The Problem Isn't the AI, It's That I Approved Them

How They Broke the Top AI Agent Benchmarks — and What That Says About My Stack

A 'perfectible' language: why the idea is beautiful and why it'll fail anyway

pnpm vs npm vs yarn vs bun: The Real Comparison Nobody Gives You in 2025

France Ditches Windows for Linux: What We're All Missing

Security as Proof of Work: Why Compliance Saves Nobody

Your Accessibility Score Is Lying to Your Face

Multi-Agent Software Development Is a Distributed Systems Problem