AWS Cloud Architect & AI Expert
Hand this to your technical lead. The store decision: one lookup this week, a person signs money, skip when a rule already works. Sample uses 9 Gateway tools and is not a client case study.
On June 15, 2026 Grok 4.3 GA on Bedrock Mantle at $1.25/$2.50 per 1M. On August 19, 2026 Grok 4.6 added Converse + US Geo/Global CRIS at $2.00 to $2.20 / $6.00 to $6.60. On September 28, 2026 Grok 4.7 shipped at the same rates. Do not jump on name alone.
On July 30, 2026 AWS cut Bedrock on-demand prices 80% for GPT-5.6 Luna and 20% for Terra. Luna is now $0.22/$1.32 per 1M tokens. This post covers the routing math and what not to re-price overnight.
On July 24, 2026 AWS shipped Claude Opus 5 on Bedrock with 1M-token context, 128K max output, and ZDR by default. This post covers the Bedrock vs Claude Platform routing call and a lane-by-lane rollout plan.
On 2026-07-16 AWS put PostgreSQL 19 Beta 2 in RDS Preview. Our first-party REPACK CONCURRENTLY run cut an 873 MB bloated table to 429 MB in 4.5s with 60/60 concurrent SELECTs succeeding.
Quick Suite GA October 2025; Q Business and Kendra freeze new customers after July 30, 2026. A 420-seat professional-services firm paused a $38K/year Kendra quote and cut evaluation from 9 weeks to 3 by starting on Quick instead.
For an on-prem MySQL estate (~1.1 TB, 340 tables), DMS full load + CDC into Aurora MySQL cut planned downtime from 6h to 18 min, though LOB misconfiguration added 11 hours of rework before cutover.
For a mid-market retailer (~2.4M SKUs, checkout p95 1.9s → 420ms), DynamoDB cart + Step Functions checkout + conditional inventory cuts held 2.1k checkout RPS, after a cache-as-inventory mistake oversold 380 units in 12 minutes.
For a retailer DW exit (~42 TB logical, 18 nightly Glue jobs), Aurora→Redshift zero-ETL retired 11 jobs and cut dashboard freshness from 6h to ~15 min. Concurrency scaling on zero-ETL (March 2026) absorbed Monday open spikes.
For a .NET monolith (~220 endpoints, ~$28k/mo EC2), strangling 14 endpoints to Lambda cut p95 on those routes from 1.4s to 210ms. An early LMI move on spiky traffic wasted ~$3.2k in idle capacity provider hours.
February 2026 server-side tool execution via AgentCore Gateway + Responses API cut median tool-round-trip latency ~180 ms → ~95 ms on a 12-tool SaaS CRM assistant by removing the client orchestration hop.
For a multi-district LMS (~52k concurrent exam submissions in 30 min), SQS decouple + Aurora Serverless v2 held p95 submit at 410 ms, and off-peak monthly compute dropped 71% vs always-on EC2.
Our articles share what we know. Our consulting engagements apply that knowledge to your specific AWS environment.
