Tech Meridian ← ENTITY INDEX
PROMY MERIDIAN RU

COMPANY · ENTITY #12514

LangGraph

Related event timeline, sources and context from the news index.

EVENT TIMELINE

8

CODING · 1 SOURCE · AWS Machine Learning

Amazon Bedrock AgentCore: three strategies to cut token costs and latency for large JSON agent workloads

Amazon describes three composable strategies used in a production Flightscape agent (Training Schedule Optimization) on Amazon Bedrock AgentCore to avoid passing multi‑MB JSON payloads inline through LLM context. By hydrating/dehydrating session state to AgentCore memory and S3, storing oversized tool results in S3 with compact handles and on‑demand pagination (get_result_page_tool), and summarizing large artifacts, they report up to 71% token reduction and up to 80% average latency reduction on a representative 23 MB payload; companion code is on GitHub.

6.0

COMPANIES · 1 SOURCE · AWS Machine Learning

Duco rebuilds agentic AI stack on Amazon Bedrock and AgentCore

Duco, an operations platform for financial services that processes 20 billion transactions monthly for 200+ clients, migrated its agentic AI stack to Amazon Bedrock and Amazon Bedrock AgentCore. The move preserved tenant data isolation and compliance while giving Duco managed runtimes, compatibility with existing LangChain/LangGraph code, and access to multiple foundation models to reduce deployment time and onboarding costs.

6.0

COMPANIES · 1 SOURCE · InfoQ AI, ML & Data Engineering

Uber’s Legal Redlining Agent cuts contract review time by ~20% using RAG and rules

Uber says its Legal Redlining Agent reduced average contract review time by more than 20% and achieved 91% accuracy in AI-generated decisions. The Microsoft Word add-in combines retrieval-augmented generation with historical lawyer feedback, deterministic rules, a React/TypeScript front end, a Python backend orchestrated via LangGraph, OpenSearch vector stores, and Redis session state; lawyers retain final review and approval responsibility.

7.0

COMPANIES · 1 SOURCE · AWS Machine Learning

Amazon Bedrock AgentCore gains Web Search API; LangChain/LangGraph integration

AWS announced a Web Search API for Amazon Bedrock AgentCore that returns signed, citeable web results from Amazon’s own index and supports domain allow/block lists and date-range filtering. LangChain’s AgentCore web search tool now calls this API and the post demonstrates a LangGraph daily dependency-watcher example using Anthropic Claude in Bedrock and AgentCore Gateway for multi-team deployment.

8.0

CODING · 1 SOURCE · AWS Machine Learning

Amazon introduces UAEF, a framework-agnostic closed-loop evaluator for AI agents

UAEF (Universal Agent Evaluation Framework) is a framework-agnostic system that normalizes agent execution traces from multiple SDKs (e.g., LangChain, LangGraph, Strands, Amazon Bedrock AgentCore) into a canonical AgentTrace, scores runs with 36 built-in metrics across seven quality dimensions, and persists experiments in Amazon DynamoDB and S3 to enable version comparisons and reports. Customers across 12 organizations in telecom, healthcare, and financial services have used UAEF and the project includes adapters, PII screening, a playbook for metrics/datasets, and runnable notebooks for supported frameworks.

6.0

RESEARCH · 1 SOURCE · arXiv cs.AI

Permission to Retry: Study of authorization and recovery in AI agent workflows

A new arXiv paper (arXiv:2610.31916v1) studies execution-time reliability for tool-using AI agents by separating recovery decisions from language-model behavior. The authors define distinct obligations (checkpoint state, permission to dispatch, evidence of external effect, permission to reconcile) and report a finite conformance study that ran 336 histories across LangGraph and Temporal hosts (14 schedules, three provider contracts, four application policies), with retained-ledger reconciliation passing all 42 checked histories and varied success rates for cached retry (30), revalidation before retry (36) and stable-key retry (31); a 16-case component study exercises InfinityOS (Arya.ai) recovery and settlement code over repository fakes.

5.0

COMPANIES · 1 SOURCE · AWS Machine Learning

Eon uses Amazon Bedrock AgentCore Gateway to centralize multi-cloud MCP access

Eon implemented a centralized authenticated gateway using Amazon Bedrock AgentCore (Gateway, Runtime, Policy, Identity) to replace per-tool credentials and local MCP processes, giving engineers and autonomous agents unified access to investigation tooling across AWS, GCP, and Azure. The change reduced developer onboarding for investigation tooling from about two days to roughly two minutes and lets Eon's autonomous SRE agent (Claude on Bedrock) access the same tools with separate identities and permissions.

6.0

COMPANIES · 1 SOURCE · InfoQ AI, ML & Data Engineering

DigitalOcean launches Managed Agents in public preview

DigitalOcean launched DigitalOcean Managed Agents in public preview, a managed cloud infrastructure layer for AI agents that combines a microVM-based Harness Runtime for persistent, isolated agent sessions and an Action Gateway that centralizes and governs tool access (via an MCP endpoint) with connectors to many services; features include pause-when-idle sessions, parallel execution, security-hardened compute/storage, and access to over 16,000 tools.

6.0