MCP vs REST: what changes when agents call your API
MCP does not replace REST. It changes who the caller is. What that shift means for interface design, auth, error messages, and operations.
Elva Blog5 min read
engineering
1,108 stories
The missing feedback loop for software factories
Warp Factories make it easy to automate your SDLC with coding agents. The harder challenge is building a factory that learns from feedback and improves with your team.
Warp Blog7 min read
Kubernetes 1.37 Breaking Changes: What to Check Before You Upgrade
Kubernetes 1.37 removes 25 feature gates, an alpha scheduling API, and 18 kubelet flags, and starts the nftables transition. What breaks, and how to check your cluster.
Radar Blog12 min read
How We Fixed Our Cloudflare Durable Objects Memory Exceeded Errors
Cloudflare Durable Objects run in V8 isolates with a hard 128 MB limit. Ours idled at ~140 MB and was reset ~300 times a day. 130 MB of zod schemas were built at module load, most of them by code that was never called.
Polylane Blog18 min read
How Devin rebuilt its renderer
The largest Devin sessions took 20+ seconds to load and crawled afterwards. We built a new chat renderer — skeleton outlines, island loading, and scroll anchoring — so even sessions with hundreds of thousands of events load fast and scroll smoothly.
Windsurf Editor (Codeium)7 min read
Source code as primary context for coding agents
Coding agents should inspect source code directly, use deterministic search for retrieval, and reserve model reasoning for understanding and decisions.
GitHits beta 0.9 Blog4 min read
Approve refunds over $500: the four things that decide whether a threshold holds
Every guide says auto-approve under $100 and escalate above it. That is the easy part. Four properties decide whether an approval threshold survives contact with production.
Pushary Blog8 min read
The AI Model Plateau: Why Your Infrastructure Matters More Than the Next Release
AI model routing and evaluation now decide your production results. Learn how companies are adapting to make the most of the highly competitive LLM landscape.
Ito Blog14 min read
Introducing scripting workflows for smarter debugging
We have recently launched a new workflow step for introspecting fatal app issues with a twist— rather than using predefined building blocks, the actual logic is written as a script, to capture exactly what matters for your team.
bitdrift Blog8 min read
Graph engineering is missing a graph
Graph engineering stops at the diagram. The harder half is the context graph underneath: whether relations, vectors and memory share one consistency boundary.
SurrealDB Blog20 min read
Everything Is Green and Nothing Works: Network Reachability in Radar
Pods Ready, Service green, curl times out. Radar's Reachability tab traces the declared network path and live-probes it, naming the first hop that breaks.
Radar Blog9 min read
Testing known time in Agent Memory on a 2,200-year corpus
How to use asOf in Agent Memory to allow gating of spoilers to ensure that spoiler-filled memory, even if already ingested, does not show up if it should not.
SurrealDB Blog8 min read
With software alone, one B200 beats the LPU and gets close to Cerebras
Stock SGLang served gpt-oss-120b at 411 tokens per second on one rented B200. Four settings took the same card and the same weights to 1,366, and single prompts to 2,215. No kernel written, nothing recompiled.
RunInfra Blog17 min read
Integrating Ultima VII with SurrealDB Agent Memory: the testing
How Ultima VII was integrated with SurrealDB Agent Memory via Exult code changes to emit events and a Rust app using egui between the game and Agent Memory.
SurrealDB Blog9 min read
When Software Becomes the User
Explore how scheduled jobs, backend agents, and services can call MCP capabilities, with scoped authorization and no need for a person to start a chat.
Noodle Seed Blog7 min read
Agent memory needs three clocks: tri-temporal belief history
Why tri-temporal belief history is needed for agent memory to properly simulate human memory and how to work with SurrealDB Agent Memory to best use it.
SurrealDB Blog5 min read
AI adoption starts with truth
The semantic layer is the foundation AI adoption is limited by trust. A user who gets burned by a confidently wrong answer will double-check the next one, eventually routing consequential work around the system entirely.
Ghostwriter Blog8 min readDropstone 1.7 Models
All three tiers changed. Heavy moves to Kimi K3, the highest-scoring open-weight model on the Artificial Analysis Intelligence Index and the first open model to clear a flagship closed one. Pro takes GLM-5.2, last cycle's paid-only Heavy, now free on every plan.
Dropstone Blog11 min read
Showing 1,008 of 1,108 stories
Load more storiesAiByts Weekly Brief
Get the top AI news and tools straight to your inbox every week.
Weekly · AI news
Join 15,000+ AI enthusiasts