All
Articles 131,250Blog Posts 136,246Tech Tutorials 33,996Research Papers 25,453News 18,664
⚡ AI Lessons
Dev.to · Pavel Espitia
2h ago
What 67 Dev.to Posts in Three Months Taught Me About Writing for Developers
Since the middle of April 2026 I have published 67 posts on dev.to. My best one has 576 views. That...
Dev.to · Pavel Espitia
1d ago
The Cheap Way to Add AI Review to CI: Small Local Models Plus Prompt Caching
I wired an AI code reviewer into CI, felt clever, and then looked at the bill after a busy week of...
Dev.to · Pavel Espitia
1w ago
Run Qwen Coder & DeepSeek Locally: The 2026 Free AI Pair-Programmer Setup
You're paying $10 to $20 a month for Copilot. You don't have to. A 2024-era laptop can run a coding...
Dev.to · Pavel Espitia
🧠 Large Language Models
⚡ AI Lesson
1w ago
Fable 5 Just Shipped: What Anthropic's Newest Model Means for Developers
On June 9, 2026, Anthropic shipped Claude Fable 5, a model in a new tier that sits above Opus. I have...
Dev.to · Pavel Espitia
1w ago
Adaptive Thinking Killed My Token Budget Code: Migrating Off budget_tokens
I had a tidy little helper that computed a thinking budget based on input size. Something like "give...
Dev.to · Pavel Espitia
1w ago
The $292M KelpDAO Bridge Hack: Why the Audit Wasn't the Problem
On April 18, 2026, attackers drained about 116,500 rsETH, roughly $292 million, from a cross-chain...
Dev.to · Pavel Espitia
1w ago
Hybrid Local + Cloud LLMs in 2026: When to Use Ollama and When to Pay for Fable
I run a local model and I pay for cloud models, and the most common question I get is "which one...
Dev.to · Pavel Espitia
🧠 Large Language Models
⚡ AI Lesson
1w ago
Writing Evals for an LLM Security Tool: How I Know It Didn't Get Worse
Every time a new model ships, I face the same question for spectr-ai: does my contract auditor get...
Dev.to · Pavel Espitia
1w ago
Prompt Caching Cut My Claude Bill by 80%: The Mistakes That Were Costing Me
I was paying full price for input tokens I was sending over and over. A large system prompt, a fixed...
Dev.to · Pavel Espitia
2w ago
Signature Replay Attacks in Solidity: The Bug That Looks Correct
Some smart contract bugs scream at you. A missing onlyOwner, a reentrancy with the state update after...
Dev.to · Pavel Espitia
🧠 Large Language Models
⚡ AI Lesson
2w ago
Structured Outputs vs Tool Use vs Prefills: Getting JSON Out of Claude in 2026
For a long time the trick to force a model into JSON was to prefill the assistant turn with an...
Dev.to · Pavel Espitia
🧠 Large Language Models
⚡ AI Lesson
2w ago
Building Multi-Tenant RLS in Supabase: Lessons From Shipping Lomi
Lomi is a SaaS for dog kennels in Colombia, and it has two kinds of tenancy at once: each kennel is a...
Dev.to · Pavel Espitia
🤖 AI Agents & Automation
⚡ AI Lesson
2w ago
I Built a Coding Agent With the Manual Tool-Use Loop. Here's What It Taught Me
Everyone reaches for the SDK's tool runner because it hides the agentic loop. I wrote the loop by...
Dev.to · Pavel Espitia
⚡ AI Lesson
2w ago
Proxy Contracts and Storage Collisions: The Upgrade That Corrupts Your State
Upgradeable contracts are everywhere in DeFi, and they hide a category of bug that has nothing to do...
Dev.to · Pavel Espitia
2w ago
Effort Levels in Practice: I Benchmarked low Through max on Real Tasks
The current Claude models give you an effort knob with five settings: low, medium, high, xhigh, max....
Dev.to · Pavel Espitia
🔧 Backend Engineering
⚡ AI Lesson
3w ago
Reading a Verified Contract You Didn't Write: A Systematic Approach
Opening a 600-line Solidity contract you have never seen is intimidating. Where do you even start?...
Dev.to · Pavel Espitia
⚡ AI Lesson
3w ago
What I Learned Submitting a Chrome Extension to the Web Store
I built Argus, an AI transaction firewall as a Chrome extension, and submitting it to the Web Store...
Dev.to · Pavel Espitia
🧠 Large Language Models
⚡ AI Lesson
3w ago
RAG for Code: Why Chunking by Function Beats Chunking by Lines
I built a retrieval system over a codebase so an LLM could answer questions about it, and my first...
Dev.to · Pavel Espitia
3w ago
The Drift Protocol Hack: A Six-Month Social Engineering Operation
On April 1, 2026, attackers drained about $285 million from Drift Protocol on Solana. It was the...
Dev.to · Pavel Espitia
🔐 Cybersecurity
⚡ AI Lesson
3w ago
Detecting Supply-Chain Malware Without Running the Code
After I got targeted by a fake-job-interview repo designed to steal my keys, I built a scanner that...
Dev.to · Pavel Espitia
🔧 Backend Engineering
⚡ AI Lesson
3w ago
Chatting With Any EVM Contract: How Scry Resolves Proxies and Unverified Bytecode
Scry lets you talk to any EVM smart contract in plain English. Point it at an address on six chains...
Dev.to · Pavel Espitia
4w ago
Foundry Fuzzing vs Invariant Testing: When Each One Finds the Bug
Foundry gives you two automated testing tools that sound similar and find very different bugs:...
Dev.to · Pavel Espitia
⚡ AI Lesson
4w ago
Streaming Claude to the Browser With Backpressure That Actually Works
Streaming LLM tokens to a browser is easy to get 80% right and surprisingly easy to get the last 20%...
Dev.to · Pavel Espitia
💻 AI-Assisted Coding
⚡ AI Lesson
1mo ago
Why I Log response.model on Every Claude Call (and You Should Too)
It is a one-line habit that has saved me more debugging time than any clever abstraction: I log which...
DeepCamp AI