92 stories
My agent hit the context ceiling last week at turn 61. The framework did what every framework does now: it called a model to summarize the conversation, waited eleven seconds, and handed back a paragraph that had quietly dropped the exact error string I needed. The task failed.
My research agent had read about 20,400 documents. It had concluded nothing. Not metaphorically. The knowledge store held roughly twenty thousand rows — papers, articles, PDFs, nightly log entries — and there was nowhere in the system that said what any of it meant.
Most AI MVP budgets I see cover the build and forget the monthly line. That line is the model bill, and at a few thousand users it can pass what the build cost. One formula gets you a usable estimate.
Originally published on the WebSpeaker blog. In short: We tested Jev, a new model for rapid decisions, as a way to speed up WebSpeaker. Jev made decisions quickly and slightly more accurately than the model we use today. Yet in every variant we tested, visitors waited longer for an answer.
Every batch job I run against LLM APIs used to cost the same at 2 a.m. as at 2 p.m. Then I started routing workloads through a unified gateway that charges different rates at different hours, and my overnight evaluation runs dropped to half the token cost — without touching a single prompt.
Originally published on my blog. This cross-post includes the setup and measured results; the linked records and configuration kit let you inspect the experiment. My RTX 3090 is a leftover from my Ethereum mining days.
Beyond One-Shot Success: How AREX-2 Teaches LLM Agents to Reflect and Persevere Current autonomous LLM agents are often evaluated by their ability to solve a task in a single pass or through a short sequence of scripted interactions.
Limits, before anything else. 21 attempts on one defect are repeated observations, not 21 independent tests. One seeded bug: an inverted authorization check in material I wrote myself. Run on 21 August 2026. One small local model. The alias local-advisor-v1:latest is local.
The AI Data Satiation Point: Why "Wild" AI Text is Breaking Chinchilla Scaling Laws As we approach the end of 2026, the internet looks fundamentally different than it did just two years ago.
Un agente LLM puede parecer muy capaz durante una demostración y volverse dificil de operar cuando empieza a llamar APIs, crear tickets o consultar datos reales. El problema no suele ser que el modelo “no sepa” la respuesta.
Open WebUI is a universal chat interface with connectivity to local or remote LLM providers, custom files, documents, and a knowledge base, with full customization of model capabilities. Application features are configured both personally for a user account and globally by admin users.
We gave our long-running agent a diary. Every cycle it would write an INNER journal entry — what it observed, what hurt, what it planned to improve. The idea was that reflection would compound into self-improvement. It didn't.
Originally published in TL;JP, a weekly, fact-checked read on what Japanese AI and tech practitioners are saying on X. I read a lot of Japanese AI posts each week so you don't have to. This week there was a pattern worth naming: almost nobody was excited about talking to a model.
Suppose you’re building a service that receives customer support tickets. Before assigning a ticket, your code needs to answer two questions: Which team should handle it? Does it need urgent attention? You could ask a language model to write an assessment and then parse its response.
Claude Code makes a lot of small judgments that don't need a full LLM call: does this edit break a rule in my CLAUDE.md, which of my installed skills fits this prompt, does this diff need a careful review or a quick pass.
If you give an LLM access to your code, you write tools. A tool is a function plus what a model needs to call it: a name, a description, and a schema for the arguments. Then you write the same tools again. The first version uses the AI SDK's tool().
Anthropic released Claude Sonnet 5.5 this week, and the main takeaway isn't a new state-of-the-art benchmark. The real story is the focus on cost-per-task for the bulk of everyday engineering work.
I built a role-play game with LLMs and expected werewolves on spaceships and in abandoned castles. Some players had a different idea. What happened I have a little Werewolf text game with LLMs. With a caveat - there is a role-play.
Showing 72 of 92 stories
Load more storiesGet the top AI news and tools straight to your inbox every week.
Weekly · AI news from Dev.to AI
Join 15,000+ AI enthusiasts
