Articles
In-depth articles on AI, web development, cloud native, and indie hacking.
Model providers may partially time out, rate-limit requests, or silently degrade quality. This article covers incident declaration criteria, UTC evidence timelines, isolation matrices, Prometheus queries, fallback thresholds, and recovery validation.
React and Vue components are tied to their frameworks. Web Components use browser-native Custom Elements, Shadow DOM, and HTML Templates for cross-framework reuse. This guide explains their APIs, tradeoffs, best use cases, cases to avoid, and how Lit improves the development experience.
A new model version cannot be released based on container health alone. This article designs a model-serving rollout process spanning offline gates, shadow requests, staged canaries, and automatic rollback, while addressing caches, sessions, and GPU warm-up.
The first step to the cloud isn't creating a VM — it's understanding networking. Covers VPC/subnets/routing/security groups/NAT: public vs. private subnets (databases shouldn't face the public internet), security groups and NACLs, NAT gateways, and a classic 'two-tier subnet + minimal exposure' architecture.
Search results do not prove that they belong to the current commit. This article uses commit watermarks, change logs, deletion and rename handling, dirty-worktree overlays, and atomic index publication to prevent coding agents from reading stale symbols.
When services run across machines, logs become scattered and hard to find during incidents. Log aggregation centralizes them for search, analysis, and alerts. This article explains the collection → transport → storage → query architecture, ELK vs. Loki tradeoffs, why structured JSON logs matter, and how to set level, sampling, and retention policies.
Routing tasks to the “strongest Agent” by role name ignores tool permissions, version compatibility, and real-time capacity. This article uses versioned capability descriptors, tiered admission control, and deadline-aware load shedding to protect interactive and critical workloads.
`useEffect + fetch` looks simple, but it is full of traps: races, duplicate requests, no cache, and no retries. This article traces React data fetching from `useEffect` to React Query/SWR and Server Components, explaining when to use each and why RSC and RQ complement rather than replace each other.
Drawing a DAG does not make it recoverable. This article separates dependency satisfaction, node attempts, artifact commits, and graph revisions, using event replay and a derived ready set to prevent missed runs, premature runs, or duplicate downstream release after a restart.
OFFSET pagination becomes painfully slow by page 1,000 because it really scans all preceding rows. This article explains OFFSET's O(N) cost, O(1) keyset pagination, compound cursor design, and when page jumps still make OFFSET necessary.
After multiple agents return results, concatenation or voting loses evidence and applicability conditions. This article designs candidate-result lineage, deterministic conflict classification, arbitration records, and atomic publication so merged conclusions can be recomputed and revoked.
When JavaScript is too slow for image and video processing, 3D, cryptography, or heavy computation, WebAssembly lets browsers run code compiled from C or Rust at near-native speed. This article explains what WASM is, how it complements JavaScript, how to use it, and where it does and does not fit.