Kubernetes Events Expire in an Hour. Your Incidents Don't.
Kubernetes events expire after one hour by default, and many of the changes that cause incidents never become events at all. How to keep history that survives.
Radar Blog11 min read
engineering
1,108 stories
MCP 2026: From Tool Protocol to Operating Layer
Explore MCP’s move toward stateless requests, guided interactions, and durable work, plus Noodle Seed’s approach to supporting different client versions.
Noodle Seed Blog10 min read
All our Agents get computers, we pay for almost none
Construct is an AI employee with a real Linux computer per user. Don't get excited - we are not paying for all of them. How we run that on Cloudflare Durable Objects, Sandboxes, and R2.
Construct Computer Blog12 min read
$0.09 and $290.12: What Actually Moves Your Inference Bill
Nine cents and $290.12 are both prices for one million output tokens. The gap is driven by accelerator, provider, concurrency, utilization, and reasoning token volume.
RunInfra Blog10 min read
Under the Hood: Serving Kimi K3
How DigitalOcean served Kimi K3 on day 0: GPU selection (B300/MI350X), vLLM tuning, and the OpenAI-compatibility fixes for tool calling, streaming, and reasoning.
DigitalOcean Blog14 min read
Serving Kimi K3 on vLLM was hard. Here is what we measured.
No released vLLM build served Kimi K3 when we started. We resolved several distinct failures on 8x B300 and measured the result, its operating crossover, and its honest limits.
RunInfra Blog10 min read
Integrating Ultima VII with SurrealDB Agent Memory: the game
Why Ultima VII through the Exult open-source game engine ended up being the perfect testing platform for SurrealDB Agent Memory
SurrealDB Blog9 min read
Stadium Summer: The Snyk Connect Fan Zone Tour
Snyk’s Fan Zone tour brought AI security workshops, networking, and friendly competition to 8 cities and 3 virtual sessions. Attendees built skills, shared ideas, and leveled up together.
AI Code Completion by DeepCode Blog3 min read
Read-Only Is Not a Safety Boundary
Two sentences get said in every meeting where a team wires an AI agent into a cluster: it's read-only, and just use the service account. Both are wrong later.
Radar Blog15 min read
Process Was Always the Bottleneck
Every week an engineer I respect tells me they're worried a coding agent is coming for their job. Engineers spend about 30% of their time writing code, and the more senior you get, the less you write. The agents took the typing.
Oats Blog7 min read
The Cloud Software Factory Build Guide
A cloud software factory is an automation loop around the software development lifecycle: triage, spec, implement, review, verify, ship, monitor. This guide walks through building one from scratch on GitHub Actions runners, one skill at a time.
Warp Blog4 min read
MTTR is the Wrong Metric for AI-Era Engineering Teams
AI tools produce 41% more bugs and 98% more pull requests. MTTR can't keep up. Here's how MTTF shifts your team from incident response to prevention.
Ito Blog14 min read
The problem with hypergrowth AI startups
Startups that have had explosive AI-driven revenue growth might be in trouble. This is a consequence of the exact same open-weight model and AI sovereignty dynamics that are driving hype and discussion across twitter right now.
Warp Blog5 min read
Why We Don't Turn Context Into a Tool
Why context delivery is a trust decision: compare injected application state with MCP tools, and separate trusted instructions from untrusted data.
Noodle Seed Blog6 min read
Query Profiling: See Where a Slow Query Spends Its Time
When a Weaviate query is slow, the first question is where the time went. Query profiling returns a per-stage, per-shard timing breakdown, making query performance issues visible.
Weaviate9 min read
Your fastest model is probably not your fastest model
Tokens per second measures how fast a model emits, not how fast it finishes. Our 13-model probe shows the gap that flips both the speed + cost leaderboards.
Ito Blog10 min read
Radar MCP vs kubectl: A Kubernetes Agent Benchmark on 50 Faults
The same model diagnosed 50 live Kubernetes faults 3x faster on average through Radar's MCP server than through raw kubectl, with slightly higher accuracy. What we measured, why it happens, and where Radar still loses.
Radar Blog9 min read
2 GitHub Actions YAML Files That Let Every Engineer Ship a Ton of Code
I opened 80 pull requests in a two-week sprint by assigning tasks to an agent and walking away from my machine. The tool is called Ivan, it's free and open source, and it runs natively in GitHub Actions.
Oats Blog14 min read
Showing 1,026 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