# Build for the Memo, Not the Demo — Shawn Chan, China Resources Holdings

Source: [AI Engineer](https://www.youtube.com/watch?v=tJFjeMBKbIY)  
Feed7 permalink: https://feed7.dev/p/build-for-the-memo-not-the-demo-shawn-chan-china-resources-holdings-0i3s3oo  
Published: 2026-07-30T02:00:06.000Z  
Trust: Source Linked (source_linked)

## Why Included

Finance agents need claim-level provenance, explicit uncertainty, consistency checks, surfaced contradictions, and logged approval. Fluent output without those controls will not survive diligence.

## Source Summary

Chan separates a five-minute demo from a decision memo assembled from conflicting filings, transcripts, and notes. His required controls are **claim-level source links**, visible separation of facts from estimates, automatic number checks, surfaced contradictions, and a logged human approval gate.

## Practical Implication

Build outputs to answer “where did this come from?” in one click. Preserve source trust levels and disagreement through retrieval and generation, and make the system refuse inconsistent figures instead of relying on a late manual review.

## Agent-Ready Context

Chan separates a five-minute demo from a decision memo assembled from conflicting filings, transcripts, and notes. His required controls are **claim-level source links**, visible separation of facts from estimates, automatic number checks, surfaced contradictions, and a logged human approval gate.

Build outputs to answer “where did this come from?” in one click. Preserve source trust levels and disagreement through retrieval and generation, and make the system refuse inconsistent figures instead of relying on a late manual review.

These recommendations come from finance diligence experience rather than a comparative evaluation. They improve auditability but do not prove that a claim is correct, and accountability still rests with the person approving the decision.

## Connected Context

Feed7 judgment across 297 accumulated Signals:

This makes auditability an output contract rather than a final review step: every claim retains provenance, facts remain distinct from estimates, contradictions stay visible, and inconsistent numbers fail before approval. It reinforces prior finance patterns that reserve numeric mutation and validation for deterministic systems, while clarifying that traceability and formal checks verify evidence or derivation—not the truth or intended meaning of a claim.

- [How Kepler Built Verifiable AI for Financial Services — Vinoo Ganesh](https://feed7.dev/p/how-kepler-built-verifiable-ai-for-financial-services-vinoo-ganesh-0yqmhy7) — Kepler’s deterministic persistence, calculation, and rejection of unverifiable values directly implements the memo’s automatic number checks and refusal of inconsistent figures.
- [CausalForge: A Formally Grounded, Self-Improving Agentic Framework for Automated Research in Causal Inference](https://feed7.dev/p/2607-22511v1-0mgsdh3) — Its distinction between proving a derivation and proving that it matches the intended claim mirrors the limit of claim-level provenance: auditability does not establish correctness.
- [What Does Done Even Mean? Agents and Paperclip's Liveness Model - Dotta, Paperclip](https://feed7.dev/p/what-does-done-even-mean-agents-and-paperclip-s-liveness-model-dotta-pap-0lx8wfc) — Its completion model reinforces that sourced evidence and verification still require explicit human authority, residual-risk assessment, and ownership.
- [AI tools for Forward Deployed Engineering — Vasuman Moza, Varick Agents](https://feed7.dev/p/ai-tools-for-forward-deployed-engineering-vasuman-moza-varick-agents-12kjg79) — Its process-first approach provides the prerequisite for deciding which diligence steps, exceptions, and approvals should be encoded rather than automated indiscriminately.

## Context Map

- Layer: agent
- Domains: research, data
- Topics: agent-reliability, retrieval, harness-engineering

## Uncertainty

- These recommendations come from finance diligence experience rather than a comparative evaluation. They improve auditability but do not prove that a claim is correct, and accountability still rests with the person approving the decision.

## Agent Instruction

Use this item as source-backed context. Do not invent claims beyond the linked source. If this item conflicts with another source, call out the conflict.