Masli Yahaya
Technical Director @ AINNA | CTO
30+ years across engineering, enterprise IT, automation and systems development, now leading the technical architecture behind AINNA NeuralOps.
Comparing AI Agent repetition vs Detached System with Smart Routing for 100 SME bank statement sets.
This is a concrete demonstration of the Smart Routing + Detached Systems layer in the AINNA Efficiency Flywheel. 87% is an internal benchmark on this workload.
100 SMEs upload 100 different sets of bank statements. The goal is to generate accurate financial statements for all SMEs.
Small & Medium Enterprises
Different Bank Formats
Generated Reports
The AI agent reads, understands, classifies, calculates, validates and generates each report individually repeated 100 times.
AI is used once to build a reusable detached system. Bank statements are segmented and intelligently routed with minimal context.
Fewer tokens. Same outcome. Much smarter system design.
This study demonstrates the Smart Routing + Detached Systems layer of the Efficiency Flywheel. 87% token reduction is an internal benchmark on the tested bank-statement workload. Full loop: Segmentation → Smart Routing → Distillation → Detached Systems → Private Infrastructure. See LLM Strategies and Model Distillation.
Transparent model inputs for investor due diligence. This study supports the unit-economics framework in the pitch deck, not a company-reported revenue claim.
| Parameter | Base Case | Best Case | Worst Case |
|---|---|---|---|
| SMEs processed | 100 | 100 | 100 |
| Statements per SME | 12 / year | 12 / year | 12 / year |
| Avg tokens / statement (Approach 1) | ~8,500 | ~6,000 | ~12,000 |
| Avg tokens / statement (Approach 2) | ~1,100 | ~800 | ~1,600 |
| Token price (external API) | $0.002 / 1K | $0.0015 / 1K | $0.003 / 1K |
| Annual savings (100 SMEs) | ~$5,200 | ~$7,800 | ~$3,100 |
Study date: July 2026 · Local inference API fee assumed RM 0/token (GPU infrastructure amortization tracked separately)
This study is supporting evidence for the RM 2M corporate seed round. The 87% token reduction is an internal benchmark on the tested bank-statement workload, modelled with stated assumptions, not an externally audited result.
* Internal benchmark on a modelled workload. Independent audit available on investor request.
Lower GPU compute and data center load
Significant during heatwaves when cooling demand spikes
Structured systems make automation viable for smaller businesses
Rules + selective AI + human review layer
Masli Yahaya
Technical Director @ AINNA | CTO
30+ years across engineering, enterprise IT, automation and systems development, now leading the technical architecture behind AINNA NeuralOps.
Basic Package
Special SME traction programme by AINNA.
Limited starter website offer. AI usage, hosting, maintenance and custom integrations depend on the confirmed scope.
View SME Offer →
