Cost & PricingSetting an AI Budget for a Small Engineering Team
A bottom-up method for budgeting model spend on a team of five to twenty, including the buffer to hold, the caps to set, and the alerts that matter.
ReadNotes on models, agents, and what inference costs
Practical writing for developers building with large language models — how they work, how to pick one, and how to keep the bill predictable.
Cost & Pricing
ModelsSmall vs Large Models: Why Parameter Count Stopped Meaning Much
Mixture-of-experts split model size into total and active parameters, and only one of them predicts your bill. How to think about size when picking a model.
Read
FundamentalsSpeculative Decoding: Faster Output, Identical Tokens
Speculative decoding drafts several tokens cheaply and verifies them in one pass, cutting latency without changing the output distribution. How it works.
Read
ModelsStop Reading Benchmarks. Build a 20-Task Eval Instead.
Public leaderboards tell you almost nothing about your codebase. Here is how to build a private eval set in an afternoon and actually know which model to use.
Read
GuidesStreaming LLM Responses: SSE, Buffering, and Why Output Stalls
Streaming is what makes an AI feature feel fast. It is also where proxies, buffers and framework defaults quietly break things. A practical guide.
Read
AI AgentsStructured Outputs: Getting JSON You Can Actually Parse
Asking a model for JSON and hoping is a bug waiting to happen. Schema-enforced outputs, and the validation you still need around them.
Read
AI AgentsSubagents and Delegation: When Splitting an Agent Helps
Subagents buy parallelism and a clean context window, and cost you shared understanding. Here is when that trade is worth making, and when it is not.
Read
FundamentalsSystem Prompts: What They Actually Do and How to Write One
A system prompt is not a control panel and not a security boundary. Here is what it really is, where it sits in the instruction hierarchy, and how to write one.
Read
FundamentalsTemperature, Top-p and Sampling: What the Knobs Really Do
Temperature is not a creativity dial and top-p is not a quality setting. Here is what each sampling parameter changes mathematically, and how to set them for real work.
Read
ModelsThe Best Model for Refactoring Is the One That Stays in Scope
Refactoring is a constraint problem, not a generation problem. What the task actually demands from a model, and how to measure it on your own repo.
Read
Cost & PricingThe Hidden Costs of AI Coding Tools
The model invoice is the smallest line item. Here is the full list of what agentic coding actually costs, and a method for pricing each part yourself.
Read
Cost & PricingToken Accounting: Explaining AI Costs to Finance
Finance teams need cost drivers, allocation and controls, not a lecture on transformers. Here is how to translate token usage into terms a budget owner can act on.
ReadShowing 709–720 of 731 articles