Fetching from the wire…
01
02
03
04
05
06
07
08
Public story · 2026-09-20 · high
AWS moved Amazon Connect Talent from April preview to general availability on September 17, 2026.
Why now: This entered the 2026-09-20 research corpus from AWS (primary; corroborated by TechTarget and Unite.AI) and has enough source evidence to become a public Rabbit Hole story.
AWS moved Amazon Connect Talent from April preview to general availability on September 17, 2026. Recruiters configure competency criteria and questions per role; AI agents then run the interviews and assessments on demand, returning scored candidate summaries with transcripts and per-competency evidence rather than a single opaque score. AWS built it on Amazon's own internal hiring systems, the same playbook it used to turn in-house contact center tech into Amazon Connect, which puts direct pressure on HireVue and the screening tier of Greenhouse and Workday.
Each link below shares sources, entities, or timing with this story.
Working paper CES-26-56 by Cody Orr, Lee C. Tucker and Lawrence Warren uses administrative records, not a survey and not a vendor benchmark, to track college graduates since ChatGPT shipped in late 2022 (US Census Bureau). The most AI-exposed decile of majors saw initial emplo...
CloudWatch Coding Agent Insights tracks tokens, cost, and cache hit rate for Claude Code, Codex, and Copilot without extra setup, per AWS.
The six open-source skills take a Hugging Face model reference and return a production real-time SageMaker endpoint, picking the serving container, wiring autoscaling and CloudWatch alarms, and verifying the result (AWS ML Blog). AWS choosing skills as the integration point fo...
The Kubernetes-native GPU-aware routing add-on for EKS scores pods on KV cache utilization, queue depth, whether the requested LoRA adapter is already loaded, prefix cache hit probability and active request count. AWS reports up to 97% lower time-to-first-token on mixed GPU fl...
AWS announced it September 18, describing it as the first open model at that scale, with native vision and about 2.5x the scaling efficiency of Kimi K2. It's the first open-weight model on Bedrock to support explicit prompt caching. Ships as both a US geographic profile (us.mo...
Opt in with platformVersion=V2. The old runtime ranged from 5.4 to nearly 30 seconds as image size grew from 200MB to 2GB; the new one holds near 2 seconds across that range. It also reclaims memory when a session goes cold instead of billing peak memory for the session's life...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.