TrendPulse
🇬🇧 English 🇨🇳 中文
🔖 Saved RSS
🌍 Dev.to

What an Unused AI Feature Actually Costs

An AI feature that nobody uses is not free. It has three separate cost lines, only one of which stops when usage does, and the largest of the three is usually the one that is never written down. Here is how to total them with your own numbers, and what could have been known before the build started. Three cost lines, not one The usual accounting for an unadopted feature is engineering time, treated as sunk and therefore ignored. That undercounts badly, because two of the three li...
Tech
🌍 Dev.to

Feature Stores and the Problem They Solve

A feature store exists because the same feature gets computed twice — once in a batch job for training and once in a service for scoring — and the two definitions drift apart. This page shows one feature breaking, in eight lines, and the join that stops it. One feature, two definitions The feature is avg_order_value_30d . It appears in a fraud model. The training set is built in the warehouse: -- training pipeline, run nightly over history SELECT o . order_id ,...
Tech
🌍 Dev.to

Federated Learning, Practically: The Protocol and What It Does Not Protect

Federated learning trains a shared model across many devices without collecting their data on a server. That is a real and useful property. It is also routinely stated as though it were three properties, and the other two — that the server learns nothing about individuals, and that the resulting model cannot leak training data — are not true unless you add machinery that costs something. The protocol, step by step The canonical algorithm is federated averaging. One round looks li...
Tech
🌍 Dev.to

A SaaS App Eval Ledger for Direct API Token Cost, Fallbacks, and a Unified Key

Short answer: don't choose an LLM API for a SaaS app from a public token-price table alone; replay a representative eval set through direct OpenAI, direct Claude API, and an OpenRouter-style routing path, then choose the smallest integration whose measured quality, billed usage, retry behavior, and recovery policy meet your release constraint. The cheapest path is workload-dependent. A low input rate can hide engineering overhead, while a chatty agent can turn one user action into several mod...
Tech
🌍 Dev.to

Few-Shot Prompting: How Many Examples Is Enough?

“How many examples” is two questions wearing one coat. The cost of another example is arithmetic you can do exactly. The benefit depends on what you believe examples are for — and the published evidence on that is more surprising than the advice built on it. The cost side, exactly Examples are input tokens on every single call, forever. Take a classification prompt whose examples run about 180 tokens each (input plus label plus the delimiters around them): shots prompt ...
Tech
🌍 Dev.to

Filtering and Vector Search in One Query

Add WHERE tenant_id = 42 to a working vector query and one of two things happens: it gets slow, or it returns fewer rows than you asked for. Both are the same underlying fact — an approximate index and a filter cannot be applied at the same time — and the selectivity at which it becomes fatal is calculable in advance. Why the two do not compose A B-tree index and a GIN index can be combined: Postgres builds a bitmap from each and ANDs them. An HNSW index cannot participate in t...
Tech
🌍 Dev.to

Fine-Tuning an Embedding Model on Your Own Data

Fine-tuning an embedding model is unusually cheap — a base model of a hundred million parameters, fifty thousand pairs and an hour on one GPU. What is not cheap is everything that happens after: a fine-tuned model is a new model, and every vector you have stored becomes incompatible on the day you deploy it. When this is worth doing Fine-tuning pays when your domain’s notion of relevance differs from the general one the base model learned. Concretely: your users search with vocab...
Tech
🌍 Dev.to

The Real Cost of a Fine-Tune, End to End

Almost every fine-tuning cost estimate is a GPU-hour calculation, and the GPU hours are usually the smallest of the four terms. This page is deliberately parametric: prices in this field move faster than a page can be revised, so what is offered here is the structure and the derivations, with every figure marked as an assumption you replace. The four terms ONE-OFF, per attempt (and you will make more than one attempt) 1. data curation hours, labelling, or teacher ge...
Tech
🌍 Dev.to

Building a Fine-Tuning Dataset From Production Logs

Production logs are the best training data you will ever have and the most dangerous. They are real inputs in the real distribution — and they are full of your current model’s outputs, which is exactly what you must not train on. Selection is the entire job A log dump is not a dataset. Ninety percent of production traffic is the model already doing the right thing, and training on it teaches nothing you do not already have. The examples worth including are the ones carrying infor...
Tech
🌍 Dev.to

How Much Data Do You Need to Fine-Tune?

The honest answer is between 500 and 500,000, and which end you are at depends almost entirely on whether you are teaching behaviour or coverage. Fortunately the question is cheap to answer empirically, and the procedure at the end of this page is worth more than any number. Why the number is unanswerable as asked “How many examples” conflates two quantities that behave completely differently. One is how many examples it takes for the model to learn the pattern — the output sha...
Tech
🌍 Dev.to

Fine-Tuning Failures: The Modes and How to Tell Them Apart

The dangerous fine-tuning failures are not the ones that raise an exception. They are the ones where the run completes, the loss curve looks plausible and the model is quietly wrong — and the way to catch those is to know what each failure mode looks like from the outside. Failures that crash Out of memory partway through, not at the start A run that survives a thousand steps and then raises CUDA out of memory did not gradually leak. It met the longest sequence in ...
Tech
🌍 Dev.to

Licensing: What You May Legally Fine-Tune and Deploy

This page is not legal advice, and nothing in it is a substitute for reading the licence attached to the specific checkpoint you are using or for asking a qualified lawyer. Licence texts are revised, differ between releases in the same family, and sometimes differ between sizes in the same release. What follows is a map of what to read and which clauses tend to matter. Four separate licences, not one “Can I use this model commercially?” is four questions wearing one coat. Each la...
Tech
HackerNews

Water system controllers don't belong on the internet, says ex-NSA chief

Article URL: https://www.theregister.com/security/2026/08/07/water-system-controllers-dont-belong-on-the-internet-says-ex-nsa-chief-after-suspected-iran-attacks/5285070 Comments URL: https://news.ycombinator.com/item?id=49216362 Points: 94 # Comments: 51
Tech
🌍 TheVerge

Birdfy’s smart bird feeder is on sale for just $60

The Birdfy Feeder Rookie is a good option if you’re new to birdwatching or simply don’t want to spend a lot on a smart feeder, and several configurations are on sale. The standard model is down to $59.99 ($60 off) at Amazon, which is close to its lowest price. It includes seven days of access […]
Tech