---
title: "Agentic AI"
description: "26 posts about Agentic AI, summarised, each linking to the original."
---

# Agentic AI
> 26 posts about Agentic AI, summarised, each linking to the original.

## Articles

### [Use Your GitHub Copilot Plan in JetBrains Air – in Your IDE and in the Cloud](https://yomu.fyi/post/use-your-github-copilot-plan-in-jetbrains-air-in-your-ide-and-in-the-c.md)
- Company: [Jetbrains](https://yomu.fyi/company/jetbrains.md)
- Author: Vladimir Gromozdin
- Published: Nov 3, 2026
- Type: Announcement

GitHub and JetBrains have integrated GitHub Copilot into JetBrains Air, an agentic development environment that lets users hand coding tasks to agents and review changes in one place. Users can select GitHub Copilot as the agent, running locally in the JetBrains IDE via the Agent Client Protocol (ACP) or in the cloud with Air Teams. Tasks use the user’s existing GitHub Copilot subscription, not JetBrains AI credits, and run the official GitHub Copilot CLI. Cloud tasks execute on isolated virtual machines, allowing Copilot to continue working after the user closes their laptop, with all traffic proxied and no state carried between tasks. Air also supports automations, such as scheduled dependency updates or pull request reviews, using the same Copilot plan, though users must monitor their own credit usage on GitHub.


### [Financial Crime Mitigation with MongoDB / Part IV: Agentic AI for Case Management](https://yomu.fyi/post/financial-crime-mitigation-with-mongodb-part-iv-agentic-ai-for-case-ma.md)
- Company: [MongoDB](https://yomu.fyi/company/mongodb.md)
- Published: Nov 3, 2026
- Type: Explainer

The article describes an agentic AI pipeline for enhanced due diligence (EDD) in financial crime case management, built with LangGraph and MongoDB. It aims to replace manual, rule-based AML workflows with specialized agents that triage suspicious activity, gather evidence, and draft Suspicious Activity Report (SAR) narratives. The pipeline uses a team of agents for notification, data gathering, case analysis, SAR authorship, and compliance quality assurance. A human-in-the-loop pattern ensures each decision can be traced for auditability. MongoDB serves as the durable state backbone, while LangGraph handles orchestration and dynamic routing. The author emphasizes that enterprise agentic AI must centralize state in a persistent store rather than rely on the agent as the source of truth.


### [Beyond Infrastructure: The New Era of CIAM Innovation](https://yomu.fyi/post/beyond-infrastructure-the-new-era-of-ciam-innovation.md)
- Company: [Auth0](https://yomu.fyi/company/auth0.md)
- Author: Auth0 Team
- Published: Nov 2, 2026
- Type: Announcement

Okta was named a Leader in the Gartner Magic Quadrant for Customer Identity and Access Management, with identity framed as a growth multiplier rather than only infrastructure. Auth0 addresses the tension between faster customer experiences and expanding risk across apps, channels, and AI interactions. Its developer tooling includes serverless Actions, Forms, Flows, APIs, SDKs, quickstarts, an integration marketplace, and deployment tooling, with Experiment Center for testing authentication journeys. Conversion features include Universal Login, Embedded Login, Passwordless, Social Login with more than 75 connections, and Anonymous Sessions that preserve pre-login context. Security and agentic capabilities include Bot Detection, Adaptive MFA, Credential Guard, Agent as Principal, Auth for MCP, Auth for UCP, Cross App Access in early access, Token Vault, and human-in-the-loop authorization; Tenancy-as-a-Service and B2B Connect support tenant isolation and partner federation without a full identity-platform migration.


### [Wealth Management Client Onboarding Needs More Than AI](https://yomu.fyi/post/wealth-management-client-onboarding-needs-more-than-ai.md)
- Company: [Databricks](https://yomu.fyi/company/databricks.md)
- Author: Andrea DeSosa, Amit Kumar Jha, Siddharth Nahar
- Published: Oct 30, 2026
- Type: Problem & solution

Wealth management client onboarding stalls at handoffs between CRM, KYC, screening, custody, and workflow systems, even when individual tasks complete. A 2026 Capgemini report found 60% of 144 wealth executives lacked a unified client view, and 41% of relationship manager time goes to operational work. AI accelerates extraction, screening, and routing but does not resolve fragmented case state, evidence, and decision history across systems. The post introduces governed case context: a versioned, source-linked view connecting evidence, control results, exceptions, agent activity, and decisions. On the Databricks platform, a governed supervisor agent uses this context to assign work, flag conflicts, and call in human reviewers while maintaining auditability.


### [Build Specialized Agents on a Stack You Control with NVIDIA NemoClaw](https://yomu.fyi/post/build-specialized-agents-on-a-stack-you-control-with-nvidia-nemoclaw.md)
- Company: [NVIDIA](https://yomu.fyi/company/nvidia-developer-blog.md)
- Author: Nicola Sessions
- Published: Oct 30, 2026
- Type: Tutorial

Building a production-grade autonomous agent requires a reasoning model, a harness that manages tools and context, domain-specific tools, and a runtime that enforces security policies. NVIDIA NemoClaw provides open blueprints that assemble these components into a stack developers can own and customize, working with harnesses such as OpenClaw, Hermes Agent, or LangChain Deep Agents. The blueprints integrate a tuned harness optimized for Nemotron models, NeMo Relay for telemetry and optimization, and OpenShell for runtime governance of file, network, and credential access. LangChain demonstrated that tuning its Deep Agents harness for Nemotron 3 Ultra achieved near-frontier performance at roughly one-tenth the cost of leading closed models. NemoClaw agents can run across cloud, data center, and local systems, and can delegate tasks to or receive tasks from other agents.


### [Agentic Smart Payment Routing with MongoDB / Part I: The Multi-Modal Blueprint](https://yomu.fyi/post/agentic-smart-payment-routing-with-mongodb-part-i-the-multi-modal-blue.md)
- Company: [MongoDB](https://yomu.fyi/company/mongodb.md)
- Published: Oct 29, 2026
- Type: Explainer

False payment declines cost global businesses over $50 billion annually, with 45% of shoppers abandoning purchases after a single checkout decline. Scripted routing rules fall short due to dimensionality explosion, unstructured error interpretation, and multi-constraint trade-offs. The proposed architecture separates an asynchronous multi-agent control plane, which continuously analyzes telemetry, simulates rerouting options, and enforces cost guardrails, from a fast execution data plane that queries deterministic JSON rules in milliseconds. Specialized agents automate a five-step optimization loop: baseline analysis, segment performance, routing simulation, cost impact, and recommendation with human approval. MongoDB stores flexible payment context documents, enables vector search for similar scenarios, and serves active rules for low-latency execution, supporting the control-execution separation.


[Newer posts](https://yomu.fyi/topic/agentic-ai.md)
