AI & Machine Learning Insights
87 articles on AI & Machine Learning from the Creuto engineering and product team: practical guides, trends, and lessons from building real software.
Latest posts
More posts
Jev use cases: where a decision model fits in real software
TypeSafe lists nineteen industries for Jev and only ten decision shapes underneath them. Here are the shapes that pay, what the smart home demo teaches about pairing with an LLM, and the two cases where a rules engine or a language model still wins.
Akash Mohapatra
Sep 22, 2026
Jev JavaScript SDK in a Node app: typed answers
The JavaScript and TypeScript side of Jev: install the SDK, build a SystemOneRequest from typed questions, read answers typed by your own criteria, and keep the API key on the server where it belongs.
Akash Mohapatra
Sep 22, 2026
Jev coding agents: what they get wrong in your code
Ask a coding agent to add Jev and it often reaches for prompt-style code the API does not accept. Here is the failure mode, the instructions that prevent it, and the five checks worth running on anything an agent generates.
Akash Mohapatra
Sep 22, 2026
OpenJev and the Jev clones: what you can and cannot self-host
Jev's weights are not published and TypeSafe documents no way to run it yourself. Here is what the projects called OpenJev actually are, what an open reproduction gives up on calibration, and the three cases where self-hosting is still right.
Akash Mohapatra
Sep 22, 2026
Jev pricing: what $0.042 per million tokens really costs
TypeSafe prices Jev at $0.042 per million input tokens with output free. We work one monthly support workload end to end, compare it against posted frontier LLM rates, and show where the 444.6x claim comes from.
Akash Mohapatra
Sep 22, 2026
Google AX agent orchestrator: run AI agents that wait
Google AX, also called Agent Executor, is an Apache 2.0 runtime that checkpoints idle AI agents and resumes them in under a second. Here is what it needs and why it is not production-stable yet.
Akash Mohapatra
Sep 22, 2026
misalignment_policy_violation: why OpenAI stopped your agent
GPT-6 Astra brought misalignment monitoring, and with it a 403 that can stop an agent mid-task. What the check looks for, which request modes block versus alert, and how to design agents that fail gracefully.
Akash Mohapatra
Sep 21, 2026
OpenAI on AWS Bedrock: GPT-6 Astra via your AWS account
GPT-6 Astra and the GPT-5.6 family now run on Amazon Bedrock at OpenAI's list price. What changes is billing, IAM-based auth and a set of Responses API features Bedrock does not support.
Akash Mohapatra
Sep 21, 2026
OpenAI computer use API: what GPT-6 Astra can automate
How the OpenAI computer use API works with GPT-6 Astra, what a screenshot loop costs at published prices, where it beats RPA or an API integration, and the sandboxing and confirmations OpenAI expects.
Akash Mohapatra
Sep 21, 2026
One Vision. One Team. One Creuto.
Whatever your challenge, we bring the expertise to build, scale, and transform your digital presence with technology that works.