Insights

What we have learned putting AI agents into production

Original pieces from KOI's leadership: how readiness is scored, how human approval is built into regulated work, how the return on one agent is calculated, what separates a pilot from production, how the implementation options compare, and how to govern agents in compliance. Each has an author, a date, and its sources.

November 6, 2026 · 4 min read

How ready is your company for AI agents? The six dimensions we score

The six dimensions KOI AI scores in an AI readiness assessment, what strong looks like on each, and the minimum a company needs for a first production agent.

By Krish Subramanian, Chief AI Officer

Read the article →

November 6, 2026 · 4 min read

Human approval architecture for AI agents in regulated work

Designing human approval for AI agents in compliance, healthcare, and finance: which actions need a gate, who approves, what is recorded, how to keep it fast.

By Krish Subramanian, Chief AI Officer

Read the article →

November 6, 2026 · 3 min read

The ROI of one AI agent, worked out in the open

A transparent ROI calculation for one AI agent: the inputs, the arithmetic, where the estimates come from, and a calculator to run it with your own numbers.

By Krish Subramanian, Chief AI Officer

Read the article →

November 6, 2026 · 3 min read

From pilot to production: a 20-point checklist for AI agents

Twenty checks that separate an AI agent pilot from production: scope, data and access, governance, testing, rollout, and what to measure after launch.

By Krish Subramanian, Chief AI Officer

Read the article →

November 6, 2026 · 4 min read

Build, buy, or hire: five ways to get AI agents into production, compared

Building AI agents in-house, buying a point tool, hiring a large consultancy, using a boutique, or working with KOI: a plain comparison on time, cost, and risk.

By Krish Subramanian, Chief AI Officer

Read the article →

November 6, 2026 · 4 min read

A governance playbook for AI agents in financial-services compliance

A governance playbook for AI agents in lender compliance: action classes, approval maps, model-risk documentation, the audit record, and what examiners ask.

By Krish Subramanian, Chief AI Officer

Read the article →
Next step

Reading is free. So is the prototype.

Describe one task on the home page and see an agent working on it within two business days.

Take the free assessment →