KTern.AI Builds SAP Agents with AgentCore
AWS ML Blog · rss · 2026-07-10
KTern.AI shared how they built an agentic AI platform for SAP transformations using Amazon Bedrock AgentCore and the Strands Agents SDK.
Key takeaways include:
- KTern.AI migrated from a self-managed container stack to AgentCore, offloading hosting, scaling, memory, tool integration, identity, and observability to the infrastructure layer
- Agents are deployed via configuration with no custom orchestration code, allowing new agents to go live in 4–6 hours
- They utilize various multi-agent patterns (swarm, workflow, graph) based on workload
- The platform leverages AgentCore memory, gateway MCP, PrivateLink, and CloudWatch for persistent context, private connectivity, and observability
Reported outcomes:
- SAP project cycles shortened by 45%
- Discovery/assessment time reduced by 60–70%
- Automated detection of 90% of Finance and Sales anomalies
- Monthly savings of 480 engineering hours
- Infrastructure costs reduced by 70% compared to a self-built stack
- Production uptime of 99.8%
More from coding & agent
- HeyGen adds a media-sourcing skill for coding agents with 75k images and 10k tracks — HeyGen · 2026-07-22
- Agent search bottlenecks are now about variance, not raw latency — rohanpaul_ai · 2026-07-22
- LangSmith adds tracing for Pipecat, LiveKit, OpenAI Realtime, and Gemini Live — LangChain · 2026-07-22
- An MCP server signs every AI agent tool call into a verifiable Merkle chain — Funky_Chicken_22 · 2026-07-22
- Annotated transcript of a Claude Code team interview is now available — trq212 · 2026-07-22
- Claude Code skill uses 10 Markdown rules to make outputs ADHD-friendly — alex_verem · 2026-07-22