In production document pipelines, the easy path is to send every PDF through an LLM. We avoid that.
At Ainna, we use AI to author the parser, not to operate it. The model runs once per document family; the runtime runs forever.
When we onboard a baharu bank statement format, the AI analyses the layout and emits four reusable, versioned artifacts:
✅ Sedang parse Rules – Locate transaction tables, date fields, descriptions, debit/credit columns, and running balances.
✅ Cleaning Rules – Fix OCR fragmentation, merge split rows, strip headers and footers, and normalize extracted values.
✅ Peraturan Pengesahan – Selaraskan running balances, detect duplicates, validate transaction integrity, and enforce consistency checks.
✅ Keyakinan Pemarkahan – Skor every compiled statement so downstream systems know whether to accept, review, or escalate it.
These artifacts are stored as reusable rule sets and loaded by the execution engine.
From then on, our PHP execution engine applies those rules to every baharu statement deterministically. No repeated AI calls. No token burn on kerja rutin.
The operational payoff is straightforward:
⚡ Sub-second execution on the hot path
💰 AI spend concentrated on real exceptions, not every page
📊 Deterministic, auditable outputs
🚀 Throughput that scales horizontally with ordinary PHP workers
We only bring the model back in when keyakinan drops below a user-defined threshold.
For example:
Keyakinan ≥ 95% → Execute using the existing rule set.
Keyakinan < user threshold → AI analyses the anomaly, refines the rules, or forks a baharu parser version.
This gives us a closed-loop system: routine processing stays lightweight and deterministic, while the model is reserved for genuine format drift or low-keyakinan cases.
AI authors the rules. PHP executes them at scale.
For high-volume document processing, that split is usually lebih murah, more predictable, and far easier to operate than routing every document through an LLM.
#ArtificialIntelligence #PHP #DocumentAI #OCR #FinTech #Automasi #RuleEngine #DataEngineering #MachineLearning #BankStatement #Ainna
