Topic Hub

AI

Agentic commerce, multi-agent orchestration, model fine-tuning, runtime governance, and production deployment in modern infrastructure.

Current note

Most AI agents never reach production. Payments is where that gets expensive

The newest published argument and its visual artifact lead the page.

Read the note

Editorial posture

The library is organized around decisions that keep recurring in payment architecture, AI systems, and Doric product work. New notes lead by publish date, while topic lanes connect related arguments without turning the index into an archive wall.

Decision frame

Recommendation, rejected paths, and the operating proof behind the call.

Visual proof

Diagrams and original artifacts stay near the argument they explain.

Source posture

Public claims stay tied to released work or cited material.

AI systems

Most AI agents never reach production. Payments is where that gets expensive

The widely cited figure is that 88 percent of AI proofs of concept never reach production. Agentic payments inherit that gap, and the teams that cross it are not winning on model selection. They are winning on payments-specific infrastructure that makes an agent idempotent, governable, and rollback-safe at the rail boundary.

Lead essay

The current Doric argument leads the index ahead of the chronological archive.

Read the note

01 / Payments

Commerce architecture

Merchant-of-record decisions, checkout ownership, billing operations, settlement, reconciliation, and risk boundaries.

02 / AI tooling

Practical AI systems

Agentic commerce, multi-agent orchestration, model fine-tuning, runtime governance, and production deployment.

03 / Build notes

From idea to system

The design, architecture, and operating decisions behind Doric Stack products as they move from prototype to public release.

Latest Notes

All

Nov 18, 2025 By Navin Agrawal AI systems 3 min read

AI agents can buy now, and fraud systems were not built for it

Stripe, OpenAI, and Google shipped protocols that let AI agents complete purchases inside a conversation. The rails are already fast enough. The problem is that fraud detection was architected around a human clicking buy on a trusted surface, and agents break that assumption.

Oct 22, 2025 By Navin Agrawal AI systems 2 min read

DeepSeek-OCR reads documents as compressed vision tokens

Payments runs on documents - checks, wires, invoices, KYC files - and the OCR that reads them chokes on handwriting and damage. DeepSeek-OCR processes a page as roughly 100 visual tokens instead of a thousand text tokens, and it is open source. The headline accuracy comes with a compression ceiling worth knowing.

Oct 14, 2025 By Navin Agrawal AI systems 2 min read

What actually breaks when you fine-tune a model

Train GPT-2 on Shakespeare and ask it the capital of France - it forgets Paris. That is catastrophic forgetting, one of several failure modes that surface the moment fine-tuning meets production, especially in regulated banking where "the model flagged it" is not an answer.