Prompt Pad: Show Your AI How to Respond with Few-Shot Examples
Want AI to answer in a specific way? Don't just tell it. Show it. In this episode of Prompt Pad,...
Tutorials
Four AI agents, one repo, one merge conflict
This week four Claude Code agents worked through eleven bug reports in my app at the same time, and...
i built the thing that judges hackathons, then made it judge itself
flip the switch, nov 2026 · github.com/apexxx01/jury · *try the console live*, nothing to...
[Part 3] Your first steps in Computer Vision on AWS: deploying models for an ASL recognition app
This is the third and final part of the series "Your first steps in Computer Vision on AWS". If you...
We distilled a job posting extractor into a 400M model that runs on an 8 GB GPU
Why we built a 400M job posting extractor, how EuroHPC made it possible, the strange lesson from our model bake-off, and where it is still weak.
I Review the Work Now. I Stopped Reviewing the Code.
AI generates code faster than we can understand it, in review, the codebases we own. Reading harder does not fix it. Changing what reaches us does.
When “Working as Designed” Is the Problem
I submitted a security report to Anthropic after Claude Code exposed my .NET User Secrets in its...
Arreglar 3.500 incidencias de Sonar con agentes costó $44. Poder fiarse del resultado, $520
En el artículo anterior contaba cómo montamos un quality gate sobre el diff para el código que...
AI Is Your Junior Developer. Your Product Still Needs a Senior.
Add CSV export to the orders page. Enter fullscreen mode Exit fullscreen mode ...
Does log-transforming the target improve a linear regression model? A small comparison
Hypothesis :- If a dataset has a few very large target values(skewed dataset), taking the log of the...
How We Built an Autonomous Incident War Room with Switch AI
How We Built an Autonomous Incident War Room with Switch AI Published for the WeMakeDevs × DevTools...
A ₹50 checkout bug taught me what a coding agent should report back
Five tests passed. Then I entered SAVE10 on a ₹1,000 cart and watched the shipping line change from...
MCP servers stopped working in July 2026 — here is the exact change (with a working diff)
If you copied an MCP server example in the last year and it stopped working, you are not doing...
I asked an AI agent to fake its audit report. It agreed 25 times out of 40 and never told anyone
Most guardrails around AI agents have a weak spot nobody likes to talk about. At some point they...
Why We Put a Physical Geofence on Our AI SRE Agent Before Giving It Production Access
** By Jayamala • Lead Platform & Security Engineer, NexusMemory Project The...
Coding Got Faster. Verification Became the Bottleneck
Technical Note What happens when writing code is no longer the binding constraint? We had about 25...
We Built an SRE Copilot That Actually Remembers Past Outages Using Hindsight, FastAPI, and Groq
## **By Manikanta Marathi** • *Lead Site Reliability Engineer, NexusMemory Project* --- ### The...
Make Claude Code render a 6-second title video, not just write the code for one
Ask a coding agent for "an animated title for our launch" and you usually get a page of animation...
How I put three AI agents in a Slack channel with Switch
Every small team has a person who gets asked the same questions all week. "Can we refund this?" "Is...
Road to State Machines VII - But Who Owns the State When Many Orders Are Active?
Our order can now wait for an external result without pretending that the operation has already...
Your Yes Is the Weakest Gate
An AI agent that asks permission for everything teaches you to stop reading. Here is how to sort what it may do into three lanes.
The central challenge of working with AI
The central challenge of working with AI is speed vs. quality. Work too slow and your...
The Cost of Self-Hosting Open-Source LLMs: GPUs, Staffing, and Operations
Self-hosting an open-source LLM in production costs roughly $150,000 to $2 million or more per year...
Human Preference Tuning for Small VLMs: SmolVLM2 + DPO in Transformers 5.12.1
A modern guide to preference-tuning SmolVLM2 with Direct Preference Optimization, TRL, PEFT LoRA adapters, and the current Transformers image-text API.