---
id: ins_kesava-mandiga-substrate-driven-intelligence
title: 'Files first, vibes never. The substrate IS the work.'
operator: Kesava Mandiga
operator_role: 'Head of PMM, JustCall (SaaS Labs)'
source_url: https://www.linkedin.com/in/k3sava/
source_type: thread
source_title: 'Kesava on substrate-driven intelligence'
source_date: 2026-04-15
captured_date: 2026-05-03
domain: [pmm, ai-native, founder-craft]
lifecycle: [substrate, synthesis]
maturity: applied
artifact_class: framework
score: { originality: 4, specificity: 5, evidence: 4, transferability: 5, source: 4 }
tier: A
related: [ins_kesava-mandiga-anti-fabrication-as-rule, ins_kesava-mandiga-workflow-collapse-not-speedup]
raw_ref: ../wiki/CLAUDE.md
---

# Files first, vibes never. The substrate IS the work.

## Claim

I don't trust working memory for anything important. Every customer interview, every competitive datapoint, every claim I might one day put in a deck, it goes into a file with a category, a date, and a source. Then synthesis is a query, not a recollection. The wiki, the codex you're reading, every research repo I run, same pattern, different scopes.

## Mechanism

Memory decays. Files don't. When the substrate is queryable, the synthesis layer becomes mechanical, which means an LLM can do most of it and you spend your judgment on the edges. Without substrate, every synthesis is a from-scratch recall, which is expensive and wrong.