Tech News

Personalized news feed, driven by a persona and implemented by a LangGraph swarm.

Redesigning source control at Anthropic

At Anthropic, coding is no longer the bottleneck to building systems or doing research. Claude is able to autonomously iterate on research loops, micro-optimize complex programs, and build new services from the ground up at a speed and scale we haven’t seen before. The bottleneck, now, is all of the developer infrastructure built for human scale: source control, builds, and CI. In May 2026, Anthropic averaged around 2,500 code changes merged per day; as of September, we merged over 30,000 per day. We don’t expect this rate of growth to slow down anytime soon; the latent demand for merging…

AI/ML / dev.to

I Built a Cozy 60 FPS AI Agent Bakery Game to Teach RAG, Guardrails, Context Caching & Multi-Agent Orchestration

What I Built Who said learning AI System Design has to be all dark cyberpunk terminal...

Languages / dev.to

I put Temporal's Rust SDK 1.0 on a conference badge

How Temporal's Rust SDK 1.0 ran on an ESP32-S3 conference badge: four compatibility patches, a trivia game, and the sleep bug that surprised a booth.

General / arstechnica.com

Gemini 4 Argon begins rolling out in Gemini app and Antigravity, but only if you pay

Google's Gemini 4 Argon model rolls out today to AI Pro ($20/mo) and AI Ultra ($99/mo) subscribers, with agentic coding features in Antigravity and AI Studio locked to the higher tier. Internally, the model reduced YouTube server load by 45% by eliminating redundant emoji rendering and uncovered critical bugs in Project Wing's drone simulator dating back to 2018. API pricing is set at $2 per million input tokens and $10 per million output tokens, with a 95% discount for cached tokens, but rates will double after November 30.

General / blog.google

Gemini 4 Argon: Built to solve your hardest problems

TBD

DevTools / dev.to

Connect Your App to Your Enterprise Traces

Codename One's OpenTelemetry integration bridges the observability gap for native-compiled Java/Kotlin apps by embedding the tracer at build time, avoiding the loss of JVM instrumentation hooks. It propagates W3C trace headers from client apps through API gateways to backend services, unifying mobile/desktop traces with enterprise observability backends. The build-time opt-in (`cn1.otel.enabled` or `@OpenTelemetry` annotation) ensures tracing code is compiled into the binary, supporting both distributed traces and metrics without runtime surprises.

AI/ML / dev.to

I spent three weekends trying to secure an AI agent. Here's everything that didn't work — and the ~600 lines that did.

A few weeks ago I watched an AI agent get owned, lie to me about it, and pass its test. Not in a...

AI/ML / dev.to

Amnesia and déjà vu: how our AI agents failed after every restart (and how we fixed it)

Every restart erased our agents' memory. Every naive relaunch made them redo work they had already done, emails included. Here's how we made them remember everything and relaunch only when it's safe.

AI/ML / theverge.com

Shopify opens its catalog up to more AI agents

Shopify now allows any AI agent to browse and purchase from its catalog of over a billion listings (both on and off Shopify storefronts) using the Shop app and Shop Pay for one-click checkout. Customers set rules like budget limits and must approve each purchase, making the feature a configurable personal shopper. This expansion, previously limited to Meta's Muse, positions Shopify in the competitive AI-driven shopping landscape alongside Google and Amazon.

AI/ML / infoq.com

Production Patterns for Agents, Evals, and LLM Costs: QCon AI New York 2026

Once AI is in production, code arrives faster than it can be reviewed, token costs rise, evals have to be good enough to act on, and agents need permissions and guardrails. QCon AI New York 2026 (December 15-16) has confirmed sessions from engineers sharing how they handle these constraints. By Artenisa Chatziou