Bedrock Platform

Multi-agent document-search copilot: one strategy per query

This technical blog post (Part 1 of 2) describes building a multi-agent chat copilot for document search and the engineering changes that fixed poor ranking quality. The author explains that v1 ran two retrieval lanes (structured metadata + semantic content) in parallel, merged their hits, and reranked the union — producing plausible-but-incorrect rankings. v2 replaces that with a single structured-output router call (Bedrock) that returns a typed plan and selects exactly one retrieval strategy per query: MetadataOnly, ContentOnly, Hybrid, NoMatch, or NeedsClarification. Reranking uses Cohere over content; metadata rows are treated as unscored results. The post also details a deterministic fallback to a non-LLM router and previews Part 2, which will cover the adaptive Hybrid path (selectivity-based) and permission gating.

Defending Agent Flows Against OWASP LLM Top 10

A developer running multiple Bedrock-backed agents on DEV Community describes a pragmatic, code-first defense posture against the OWASP Top 10 for LLM applications. The post maps each OWASP risk to implemented controls (or gaps), including per-(agent,user) rate limits, a global monthly cost circuit-breaker, model max_tokens caps, a no-tools / read-only agent design, PII regex scrubbing before model input, prompt framing with explicit delimiters and anti-injection preambles, versioned prompt registry and anti-echo rules, schema validation and grounding checks for model outputs, and an agent-level kill switch with internal keys and quota gating. The author documents which risks are covered strongly, which are partially mitigated, and which remain unbuilt (notably vector/embedding store ACLs, per-user cost caps, output PII re-scan, and egress allow-lists). Code snippets and honest failure-mode notes accompany each control.

Spot AWS Cost Anomalies Before They Break Budgets

A Dev.to guide (published 2026-06-19) explains how teams can detect AWS cost anomalies early to avoid large surprise bills. The author defines a four-signal framework (service-line growth vs traffic, unexpected region, newly non-zero usage type, and daily percentage delta >30%) and recommends streaming, near-real-time detection rather than monthly bill reviews. AWS Cost Anomaly Detection is noted as a free baseline but suffers a 24–48 hour data lag; commercial tools (e.g., CloudZero, Vantage, Datadog Cost Mgmt, ZopNight, Harness CCM, nOps) read the Cost and Usage Report stream to surface anomalies within minutes and some offer auto-remediation. The post highlights the FOCUS billing schema change that broke dashboards, outlines remediation runbook steps (tag, quarantine, incident channel, root-cause), and explains limitations such as slow-burn trends, commitment distortions, and shared-service attribution gaps.