All Content tagged with Generative AI on AWS
Innovate faster to reinvent customer experiences and applications
Content language: English
Filter content
Select tags to filter
Sort by
Sort by most recent
323 results
I am testing Amazon Nova Sonic in us-east-1 using the Bedrock Runtime InvokeModelWithBidirectionalStream API.
AWS allows me to create an application inference profile sourced from both amazon.nova-son...
Two content-block shapes are rejected by the Converse API for the GPT-5.6 family even though the underlying models accept them through the Responses API. Both look like gaps in the Converse-to-OpenAI ...
Abdullah KaramanEXPERT
published 15 hours ago0 votes35 views
AWS DevOps Agent's native code integration covers GitHub, GitLab, and Azure DevOps, but not Bitbucket at this time. This article shows how to register the Atlassian Rovo MCP server as a capability pro...
**Summary**
`google.gemma-4-31b` on Bedrock (Mantle OpenAI-compatible endpoint, eu-central-1) reliably drops digits from numbers that contain consecutive repeated digits. Numbers without repeated dig...
We are trying to use Claude Opus 5 on Amazon Bedrock from Asia Pacific (Sydney), ap-southeast-2.
The Australian inference profile is visible and active:
au.anthropic.claude-opus-5
We also tested:
...
## Summary
On `bedrock-runtime` Converse, models `zai.glm-4.7` and `zai.glm-5` intermittently
return a `toolUse` block whose `name` field is not the function name. Instead it
contains the first argum...
**Problem**
I have a Flow in Amazon Quick with an Application Action step configured to send a Slack message. Instead of executing the Slack API call, the step generates a text response saying "I don...
We are currently working with Nova Sonic to build AI agents for a healthcare application.
These agents assist patients and are designed to detect signals of discomfort (such as pain, depression, or su...
Darius MousazadehEXPERT
published a month ago0 votes315 views
This article shows how to put cost governance around Amazon Bedrock so enterprises can safely give developers access to Claude and Claude Code. It walks through a field-tested pattern: team and user-l...
ShravanEXPERT
published a month ago0 votes175 views
Organizations adopting Amazon Bedrock at scale struggle to attribute model-invocation costs to specific teams, detect LLM-native anomalies like tool-call loops, and investigate spend spikes fast enoug...
WAZEXPERT
published a month ago0 votes152 views
Part 3 of a three-part series mapping the OWASP Top 10 for Agentic Applications 2026 to AWS security controls. This final part covers the system's dynamics once agents persist state and act autonomous...
WAZEXPERT
published a month ago0 votes168 views
Part 2 of a three-part series mapping the OWASP Top 10 for Agentic Applications 2026 to AWS security controls. This part covers what the agent runs at runtime: ASI02 (tool misuse and abuse), ASI04 (ag...