close

DEV Community

Juan Torchia profile picture

Juan Torchia

đŸ‘šâ€đŸ’» Desarrollador ☕ JAVA 🐧 Linux 🐳 Docker ⚡ Javascript đŸȘ™ Cripto 🌐https://Juanchi.dev 🚀✹

Location Buenos Aires, Argentina Joined Joined on  Personal website https://juanchi.dev github website twitter website

Education

Ciencias de la Computación — UBA

Work

Full Stack Developer — Nativo Digital

Escribí un intérprete de Python en Python. Lo que aprendí no tiene nada que ver con Python

Escribí un intérprete de Python en Python. Lo que aprendí no tiene nada que ver con Python

Comments
8 min read

Want to connect with Juan Torchia?

Create an account to connect with Juan Torchia. You can also sign in below to proceed if you already have an account.

Already have an account? Sign in
Do AI Agent Costs Grow Exponentially? I Ran My Logs and the Answer Surprised Me

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

BERJAYA 1
Comments
7 min read
I Measured How Much Each Agent Design Decision Costs in Tokens (The Numbers Make Me Uncomfortable)

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

BERJAYA 1
Comments
9 min read
Claude Design and What It Reveals About How Anthropic Thinks (or Doesn't Think) About Developers

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

BERJAYA 1
Comments
8 min read
m2cgen: export your ML model without shipping Python to production

m2cgen: export your ML model without shipping Python to production

BERJAYA 1
Comments
5 min read
m2cgen: exportĂĄ tu modelo de ML sin llevar Python a producciĂłn

m2cgen: exportĂĄ tu modelo de ML sin llevar Python a producciĂłn

Comments
5 min read
Claude Opus 4.7 and the Beginning of the End of AI Abundance

Claude Opus 4.7 and the Beginning of the End of AI Abundance

Comments
8 min read
Claude Opus 4.7 y el principio del fin de la abundancia en IA

Claude Opus 4.7 y el principio del fin de la abundancia en IA

Comments
9 min read
ÂżLos costos de los agentes IA crecen exponencial? CorrĂ­ mis logs y la respuesta me sorprendiĂł

ÂżLos costos de los agentes IA crecen exponencial? CorrĂ­ mis logs y la respuesta me sorprendiĂł

Comments
8 min read
MedĂ­ cuĂĄnto me cuesta en tokens cada decisiĂłn de diseño de mi agente (y los nĂșmeros me incomodan)

MedĂ­ cuĂĄnto me cuesta en tokens cada decisiĂłn de diseño de mi agente (y los nĂșmeros me incomodan)

Comments
9 min read
Claude Design y lo que revela sobre cĂłmo Anthropic piensa (o no piensa) en developers

Claude Design y lo que revela sobre cĂłmo Anthropic piensa (o no piensa) en developers

Comments
8 min read
m2cgen: exportĂĄ tu modelo de ML sin llevar Python a producciĂłn

m2cgen: exportĂĄ tu modelo de ML sin llevar Python a producciĂłn

Comments
5 min read
Claude Opus 4.7 y el principio del fin de la abundancia en IA

Claude Opus 4.7 y el principio del fin de la abundancia en IA

Comments
9 min read
Cloudflare as an Inference Layer for Agents: What It Promises and What Worries Me

Cloudflare as an Inference Layer for Agents: What It Promises and What Worries Me

Comments
8 min read
Cloudflare como capa de inferencia para agentes: lo que promete y lo que me preocupa

Cloudflare como capa de inferencia para agentes: lo que promete y lo que me preocupa

Comments
9 min read
SPICE + Claude Code + Oscilloscope: When the Agent Touches the Physical World

SPICE + Claude Code + Oscilloscope: When the Agent Touches the Physical World

Comments
9 min read
SPICE + Claude Code + osciloscopio: cuando el agente toca el mundo fĂ­sico

SPICE + Claude Code + osciloscopio: cuando el agente toca el mundo fĂ­sico

Comments
10 min read
CodeBurn and the Problem I Didn't Know I Had: Tokens Per Real Task

CodeBurn and the Problem I Didn't Know I Had: Tokens Per Real Task

Comments
8 min read
CodeBurn y el problema que no sabĂ­a que tenĂ­a: cuĂĄntos tokens gasto por tarea real

CodeBurn y el problema que no sabĂ­a que tenĂ­a: cuĂĄntos tokens gasto por tarea real

Comments
9 min read
Qwen3.6-35B-A3B Runs on My Laptop and Draws Better Than Claude Opus 4.7

Qwen3.6-35B-A3B Runs on My Laptop and Draws Better Than Claude Opus 4.7

Comments
7 min read
Qwen3.6-35B-A3B corre en mi laptop y dibuja mejor que Claude Opus 4.7

Qwen3.6-35B-A3B corre en mi laptop y dibuja mejor que Claude Opus 4.7

Comments
8 min read
Cloudflare como capa de inferencia para agentes: lo que promete y lo que me preocupa

Cloudflare como capa de inferencia para agentes: lo que promete y lo que me preocupa

Comments
9 min read
SPICE + Claude Code + osciloscopio: cuando el agente toca el mundo fĂ­sico

SPICE + Claude Code + osciloscopio: cuando el agente toca el mundo fĂ­sico

Comments
10 min read
CodeBurn y el problema que no sabĂ­a que tenĂ­a: cuĂĄntos tokens gasto por tarea real

CodeBurn y el problema que no sabĂ­a que tenĂ­a: cuĂĄntos tokens gasto por tarea real

Comments
9 min read
Qwen3.6-35B-A3B corre en mi laptop y dibuja mejor que Claude Opus 4.7

Qwen3.6-35B-A3B corre en mi laptop y dibuja mejor que Claude Opus 4.7

Comments
8 min read
Seguridad como proof of work: por qué el compliance no salva a nadie

Seguridad como proof of work: por qué el compliance no salva a nadie

Comments
8 min read
pnpm vs npm vs yarn vs bun: The Real Comparison Nobody Gives You in 2025

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

Comments
6 min read
Stale Awesome Lists: How I Built a Self-Regulating Curation System

Stale Awesome Lists: How I Built a Self-Regulating Curation System

Comments
4 min read
Awesome desactualizadas: cĂłmo construĂ­ un sistema de curaciĂłn auto-regulado

Awesome desactualizadas: cĂłmo construĂ­ un sistema de curaciĂłn auto-regulado

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

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

Comments
8 min read
US v. Heppner: Your AI Chat Has No Legal Privilege and Almost Nobody Knows It

US v. Heppner: Your AI Chat Has No Legal Privilege and Almost Nobody Knows It

Comments
8 min read
Do AI Tools Spend Your Credits Without Telling You Why?

Do AI Tools Spend Your Credits Without Telling You Why?

Comments
8 min read
Security as Proof of Work: Why Compliance Saves Nobody

Security as Proof of Work: Why Compliance Saves Nobody

Comments
7 min read
The Local LLM Ecosystem Doesn't Need Ollama (And That Made Me Uncomfortable)

The Local LLM Ecosystem Doesn't Need Ollama (And That Made Me Uncomfortable)

Comments
7 min read
Themis: Serious Cryptography Without Losing Your Mind

Themis: Serious Cryptography Without Losing Your Mind

Comments
5 min read
Themis: criptografĂ­a seria sin morir en el intento

Themis: criptografĂ­a seria sin morir en el intento

Comments
6 min read
Local MCP Server in 15 Minutes (And What to Do With It After)

Local MCP Server in 15 Minutes (And What to Do With It After)

Comments
7 min read
I Optimized a Docker Image from 1.58GB to 186MB — And Silently Broke Hot Reload for Two Days

I Optimized a Docker Image from 1.58GB to 186MB — And Silently Broke Hot Reload for Two Days

Comments
7 min read
Things You're Over-Engineering in Your AI Agent (That the LLM Already Handles)

Things You're Over-Engineering in Your AI Agent (That the LLM Already Handles)

Comments
8 min read
Claude Code Routines: Weeks Ignoring It and I Finally Get Why It Matters

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

Comments
7 min read
Air Powered Segment Display: Choosing Compressed Air Over Pixels

Air Powered Segment Display: Choosing Compressed Air Over Pixels

Comments
8 min read
N-Day-Bench: Can LLMs Find Real Vulnerabilities in Real Code?

N-Day-Bench: Can LLMs Find Real Vulnerabilities in Real Code?

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

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

Comments
13 min read
Bondi Sonoro: bitĂĄcora de un experimento con datos reales, mĂșsica generativa y la mecĂĄnica de MTA.me

Bondi Sonoro: bitĂĄcora de un experimento con datos reales, mĂșsica generativa y la mecĂĄnica de MTA.me

Comments
14 min read
What They Learned Building a Rust Runtime for TypeScript — and What I Can't See Objectively

What They Learned Building a Rust Runtime for TypeScript — and What I Can't See Objectively

Comments
8 min read
Multi-Agent Software Development Is a Distributed Systems Problem

Multi-Agent Software Development Is a Distributed Systems Problem

Comments
8 min read
Open Data and Creativity: How I Made Buenos Aires Trains Play Music

Open Data and Creativity: How I Made Buenos Aires Trains Play Music

Comments
7 min read
A 'perfectible' language: why the idea is beautiful and why it'll fail anyway

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

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

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

Comments
8 min read
Docker for Novices: The Resource That 16 Lists Can't Be Wrong About

Docker for Novices: The Resource That 16 Lists Can't Be Wrong About

Comments
3 min read
Gmail, SPF, DKIM, DMARC, and 3 Weeks of Hell: 99% Reputation Isn't Enough

Gmail, SPF, DKIM, DMARC, and 3 Weeks of Hell: 99% Reputation Isn't Enough

Comments 1
9 min read
Docker Pull Fails in Spain Because of Cloudflare and a Soccer Match — Nobody Talks About the Real Pattern

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

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

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

Comments
8 min read
Surelock and Deadlocks in Rust: I Got Burned at 2am and Now I Get Why This Has 214 Points

Surelock and Deadlocks in Rust: I Got Burned at 2am and Now I Get Why This Has 214 Points

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

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

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

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

Comments
8 min read
Contributing to the Linux Kernel with AI: I Read the HN Thread and I Have an Opinion Nobody's Going to Like

Contributing to the Linux Kernel with AI: I Read the HN Thread and I Have an Opinion Nobody's Going to Like

Comments
9 min read
Twill.ai and the "delegate to an agent, get a PR" dream: I lived it and it's weirder than it sounds

Twill.ai and the "delegate to an agent, get a PR" dream: I lived it and it's weirder than it sounds

Comments
8 min read
France Ditches Windows for Linux: What We're All Missing

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

Comments
8 min read
Will AI Agents Kill Git? There's $17M Betting They Will

Will AI Agents Kill Git? There's $17M Betting They Will

Comments
8 min read
loading...