How to deploy an AI-generated app to production
Publishing is one click. Getting an AI-built app ready for real visitors takes a domain, search settings, a phone test and a plan for what happens after launch.
Hosting, domains and going live.
Publishing is one click. Getting an AI-built app ready for real visitors takes a domain, search settings, a phone test and a plan for what happens after launch.
An open-source repo runs 20 Claude Code agents on iOS Simulators and Android emulators at once, hunting broken flows, layout and translation bugs for about $5 a run.
A new MIT-licensed repo profiles Python, Node, Ruby and other apps, ranks the slowest functions and has a coding agent port just those to Rust, with replay checks before enabling.
MR Office has published the source of its six browser editors on GitHub under AGPL-3.0, with a hosted embed, an MCP server and npm React packages for developers.
A new MIT-licensed Python project on GitHub cuts long videos into captioned 9:16 clips using audio signals and a local LLM, with no account and no uploads.
AWS's Agent Toolkit for AWS now includes a distributed KV caching skill that lets coding agents size, set up and validate an ElastiCache cache for self-hosted LLM inference.
AWS announced the SageMaker HyperPod Global Inference Gateway, a single managed endpoint that routes inference requests across clusters and Regions with automatic failover.
browser-use 1.0.7 makes long MCP browser_exec calls reply within 50 seconds with partial output, so clients with a 60 second limit get a result instead of a timeout.
Google's walkthrough shows how Gemini Enterprise reaches internal Cloud Run MCP servers using Egress Agent Gateway, Private Service Connect, scoped DNS and dual-header auth.
phonedriver, a Rust MCP server with 304 GitHub stars in 12 days, lets Claude Code, Codex and Cursor read and tap Android and iOS screens as text, with a guard before payments.
GitLab's walkthrough installs a skill in Claude Code that configures a whole group for Duo Agent Platform, then hands issue work, code review and pipeline fixes to GitLab.
GitLab published recommended configurations pairing Claude Code, Codex, Cursor and other agents with Duo, plus a setup skill for review, CI repair and security triage.
Exa optimized embedding inference on AWS Inferentia2 and Trainium2 with NKI kernels, reporting roughly 1.5x throughput and a 47% gain in performance and cost per million documents.
GitHub CLI v2.105.0 adds gh event listen, which waits for pull request events such as reviews and check suites instead of polling in a loop. Both it and Event Streams are in preview.
Cloudflare says Clairic has left preview and Clairic-flash has arrived on Workers AI, with both extraction models getting an open-weight release.
Vercel's Sandbox Python SDK reaches 1.0, stabilising its API and matching the TypeScript version, with a breaking bytes change for anyone upgrading from 0.x.
Google AI Studio now builds full-stack Kotlin Android apps with Firestore, Google Sign-In and Gemini calls, then publishes them to your personal Play library.
SMART is an open source Python SDK and Terraform stack that deploys a WhatsApp channel layer for agents on Amazon Bedrock AgentCore, covering media, memory, scheduling and Meta templates.
AWS adds Aionic Labs' open source time-series language models to SageMaker JumpStart, with a deployment path and fine-tuning via TimeNet.
The Vue package release carries [email protected], which validates nested UI message metadata in one pass, plus new gateway access policy errors and a 7 MiB Vertex download default.
A day-old GitHub project records what your app does, then checks every piece of a Rust port against that recording. It supports six languages and connects to four coding agents.
A day-old GitHub repo packages local LLM inference optimization as an agent skill, with benchmarking scripts, ceiling maths and before-and-after numbers on named hardware.
The huawei-csl project MACaQ shipped an Apache-2.0 quantization toolkit that writes GGUF files for llama.cpp and covers text, image, video and audio models.
Cloudflare has open-sourced Kitesurf, its experimental browser engine for AI agents. Builders can inspect the code, run it locally, or deploy an instance on Cloudflare Workers.
SGLang-Neuron is now publicly available in an AWS Deep Learning Container, bringing the SGLang serving workflow and PD disaggregation to AWS Trainium on native PyTorch.
Google Cloud has expanded AI agent observability and evaluation across Data Cloud, routing telemetry from BigQuery and Looker agents into customer-owned tables, with cost caps and the Prism evaluation tool.
Prisma's official plugin puts its ORM, Postgres and Compute inside ChatGPT and Codex, so one prompt can build an app, create its database and deploy it.
AWS's ML blog walks Java developers through building autonomous agents on Amazon Bedrock with Spring AI, covering tool calling, memory and AgentCore deployment.
Cursor's cloud agents can now scaffold, host and deploy web apps with Postgres, SSO auth and admin-reviewed security checks, all inside the editor.
phongent, a new Apache-2.0 Kotlin project, lets an iPhone or Android phone send tasks to several Android devices and watch progress, messages and live screens in one place.
A new GitHub repo packages an AI Java Spring Boot workflow: 20 agent roles, layered workflow templates, runnable Spring examples and an explicit list of what is not verified.
Google AI Edge Gallery now runs speech recognition and text-to-speech fully on device, with an Android ML Kit path and a cross-platform MediaPipe and LiteRT path.
Google Cloud has made an Anthropic-managed Claude deployment option generally available, running alongside the Google-managed option at the same models and pricing.
Cursor's Agent SDK keeps an agent in a directory of files, tests which tools it calls with evals, and deploys it on Cursor's infrastructure or your own.
The smolmaxxing repo patches llama.cpp to page experts from flash, reporting a 16B model on a 1 GB TV box and an 80B model on a 2022 Android phone.
agentes.chat is an open-source, self-hosted Slack alternative where AI agents join channels and DMs with budgets, approvals and MCP tools. It hit 1,358 stars in five days.
devvm puts each project folder in its own isolated VM with the toolchain it needs and coding agents ready to run, keeping secrets on the host machine.
AWS published a CDK sample that puts NVIDIA NeMo Switchyard in front of Amazon Bedrock models, then measures coding agent cost per solved task rather than per token.
API Gateway streaming support is now in Public Preview, carrying SSE, WebSockets, gRPC and chunked responses behind the same policies as your other APIs.
The local inference library ships v1.0.0 alongside ggml, committing to semantic versioning so downstream apps can spot breaking C API and ABI changes.
Google AI Studio can now build apps on Google Maps Platform from a prompt, setting up keys and quotas automatically. Sandbox limits are 100 calls per day per API; production needs Cloud billing.