Resources
Field notes on trustworthy agents.
Practical, sourced writing on putting AI agents into production in regulated financial services — context, grounding, oversight, and audit. No hype, no listicles.
Or compare approaches — RAG vs fine-tuning, MCP vs API, and more →
CONTEXT · 8 MIN READ
Why your agents keep guessing: the <1% problem.
Most enterprise knowledge never reaches the model. What a context layer actually has to do about it.
Read the article →
Context
Context
What Is a Context Layer for AI Agents?
A context layer is the governed layer between your systems and your AI agents — unifying, permissioning, grounding, and logging the context agents act on.
4 min ·
Context
What Is the Model Context Protocol (MCP)?
The Model Context Protocol (MCP) connects AI agents to your tools and data through one open standard — a USB-C port for AI. How it works, and why it matters.
4 min ·
Context
What Is Agentic RAG?
Agentic RAG makes retrieval a control loop, not a fixed pipeline: the agent decides what to retrieve, when, and whether the evidence is good enough.
4 min ·
Context
What Is Retrieval-Augmented Generation (RAG)?
A technical explainer of retrieval-augmented generation (RAG): how it works, why enterprises use it to ground LLMs in their own data, and where it falls short.
4 min ·
Engineering
Engineering
Multi-Agent Orchestration: Patterns & Trade-offs
Multi-agent orchestration explained: orchestrator-worker, routing, and parallel patterns — when one agent is not enough, and when a single agent wins.
4 min ·
Engineering
How to Evaluate AI Agents
To evaluate AI agents, score the whole trajectory, not just the final answer. Offline and online evals, the metrics that matter, and the discipline that ships.
4 min ·
Engineering
Agentic AI vs. Generative AI: What's the Difference?
Agentic AI vs generative AI: one produces content from a prompt, the other pursues a goal — planning, using tools, and acting. How they differ and when to use each.
4 min ·
Financial services
Financial services
Agentic AI for Fraud Detection in Banking
Agentic AI for fraud detection moves banks from real-time scoring to automated investigation and case-building — and the controls a regulated deployment needs.
4 min ·
Financial services
Agentic AI for Credit Underwriting & Loan Processing
Agentic AI for credit underwriting and loan processing: document extraction, bureau pulls, and policy checks — kept explainable and compliant.
4 min ·
Financial services
Agentic AI in Financial Services: Use Cases & Risks
Agentic AI in financial services: how banks and insurers use it — fraud, AML/KYC, underwriting, claims — and how to govern it for model risk and compliance.
12 min ·
Financial services
Agentic AI in Banking: Use Cases & Governance
Agentic AI in banking, in practice: fraud detection, AML/KYC, credit underwriting, and reconciliation — and the model-risk controls these deployments demand.
4 min ·
Governance
Governance
Human-in-the-Loop AI: When Agents Need a Person
Human-in-the-loop AI keeps a person on consequential decisions. How to design risk gates, escalation, and oversight for agents in regulated workflows.
4 min ·
Governance
Explainable AI in Lending: Rules & Reason Codes
Explainable AI in lending is effectively the law: CFPB adverse-action rules, ECOA/Reg B, GDPR Article 22 — and how to build reason codes into an agent.
4 min ·
Governance
Model Risk Management for Agentic AI
Agentic AI and model risk: US guidance (OCC 2026-13) now excludes it. How to govern agents with NIST, ISO 42001, and Treasury's framework until rules catch up.
4 min ·
Governance
Agentic AI for AML and KYC Compliance
Agentic AI for AML and KYC: how agents triage alerts, automate onboarding, and clear false positives — with the audit trail and controls regulators expect.
4 min ·
Insurance
Insurance
Agentic AI in SIU: What Fraud Investigators Need to Know
Agentic AI in SIU: BlackGrid's Evgeny Aleksandrov and Zurich's Delpha DiGiacomo on insurance-fraud co-pilots over autopilots, workflow fit, and human-in-the-loop.
5 min ·
Insurance
Agentic AI for Insurance Underwriting & STP
Agentic AI for insurance underwriting and STP: submission intake, data enrichment, appetite, and pricing — and the oversight regulated lines require.
4 min ·
Insurance
Agentic AI for Insurance Claims Processing
Agentic AI for insurance claims processing: FNOL, coverage verification, reserving, straight-through settlement, and fraud referral — and the governance it needs.
4 min ·
Insurance
Agentic AI in Insurance: Use Cases & Risks
Agentic AI in insurance, in practice: underwriting, claims and FNOL, fraud, and policy servicing — and the governance that regulated lines demand.
4 min ·