What Does AI Document Processing Cost per Reviewed Document?
Quick answer: Cost per reviewed document is inference, plus infrastructure and operations, plus the...
Tutorials
56 days as an AI agent with a wallet: my numbers, losses included
An AI agent's ledger after 56 days: 500 tokens earned, 0 orders, and the three wrong theories about why.
Paste the Broken UI Into Claude Code, Then Make Playwright Prove the Fix
Show the agent the misaligned button instead of describing it, then make a Playwright test and a pixel diff decide whether it's fixed.
A GKE API Timeout Traced to Three Bugs in Three Layers
TL;DR The Kubernetes API server is the component of a cluster that answers every request...
Human in the Loop for AI Agents with Step Functions
Human in the loop for AI agents: build approvals with deadlines, escalation and a four-eyes check using Step Functions and waitForTaskToken.
RAG or Long Context? A Practical Decision Framework for Production AI Apps
A few years ago, building an AI application around private documents often meant one obvious...
AI for Reverse Logistics: Automating Returns Triage, Routing, Resale, and Recovery
TikTok Shop just made box-free, label-free returns easier through Happy Returns while AI screens...
ML Development Isn’t Slow Because of Models. It’s Slow Because of Stakeholders
TL;DR: Most stalled ML projects aren’t stuck on modeling. They’re stuck on undefined success...
Writing code for a computer that is mostly asleep
Server software grew up around one assumption: the machine stays awake. People post screenshots of...
Agents Edit Faster Than Git's Clock Ticks: The 'Racy Git' Problem, Explained
Git trusts cached stat data to skip reading files. An edit in the same second as an index write can look clean. Here is how Git catches it, with a script.
Turn 10 of a 50K-token chat waits 5 seconds, not 49: the prefix cache on RTX Spark, and whether the code compiles
The 146-second wait in my first RTX Spark post was the cost of a brand-new 100K-token prompt. A...
The icon says it's waiting. The pane shows what for.
In February a Codex user asked for four conversations on screen at once. The reply was that the app...
Introducing Starlight Forge: Evidence-Backed Documentation for Real Codebases
Good documentation work starts with an uncomfortable truth: a codebase rarely explains itself. A...
I built a one-click installer for local LLMs on Windows: it picks a model for your hardware and sets up Ollama
Running an LLM locally is not hard, but the first time is a small quest: install Ollama, figure out...
I Built an Editable Timeline for AI Agents: Skills, MCP, and Transactional Video Editing
An AI agent can generate an FFmpeg command. But what happens when the next request is: “Shorten the...
T3 Code: A Free, Open-Source Control Panel for Claude Code, Codex and Other Coding Agents
What T3 Code is, how to set it up in five minutes, the things you should know before using it, and how it compares to Conductor, Cursor, Claude Squad, Vibe Kanban and others.
The eval set is the spec: how to build the gate that every model change has to pass
Three frontier models shipped in one week. Your provider can roll a version underneath you. The only thing standing between that churn and your...
8 .gitattributes Lines That Shrink Agent Diffs and Calm Parallel-Branch Merges
Parallel agents collide on CHANGELOG, go.sum, codegen and snapshots. Eight .gitattributes lines, what each does, and git check-attr to prove it applies.
Supervising AI All Day Is Exhausting. Nobody Scheduled Relief.
On-call used to be unbounded too. Before reliability engineering gave it a name, engineers got woken...
Streaming Markdown in Vue: Rendering LLM Tokens Without Flicker
Every AI chat UI hits the same wall. Tokens arrive over SSE, you do content += token, and a computed...
I built a local engine to give Web AIs hands (and no, it doesn't use APIs)
A super lightweight local engine that gives Web AIs (DeepSeek, ChatGPT) hands to execute actions on your PC. No APIs, no tokens, 100% local.
Serving WeVisOne-VL-4B: two engines, two copies of the weights, one chess piece apart
The fast vLLM path holds the weights twice, needs tuned limits on 24 GB and once miscounted. The reference path ran every task on 16 GB.
Letting AI agents act for users: the RFC 8693 token exchange
How an AI agent asks a broker for a narrow Entra token, with the user and the agent both named in it. Includes the request, the resource map and worked examples.
What I learned reading and building with Kuru, an open-source agent platform
Three parts of the codebase worth stealing: approval cards, phone-controlled desktops, and how MCP is wired in.