Aelix Echo · Document Intelligence

>reading invoice_38274.pdf...
>extracting 14 fields...
>structured.

Documents that
read themselves.

For ops teams at $50M+ insurers, banks, and manufacturers drowning in PDFs. We extract 14+ fields per invoice in 1.4 seconds at 99.5% accuracy, with an audit hash on every page.

See live extraction

Agentic AI · Document agents · Design-partner preview

Agents that read, validate, and reconcile. A human signs the write-back.

Document intelligence is not just OCR. It is deciding what each field means, validating it against a source of truth, and reconciling it back into the system that owns it. Aelix Agents do that work end to end. They capture, extract, validate, and propose a reconciliation. A credentialed human approves anything that mutates a system of record.

Document Capture Agent.

Ingests PDFs, scans, and phone photos. Classifies the document type, runs real OCR (Tesseract by default, managed providers behind the same interface), and packages the extraction with per-field confidence.

Field Extraction Agent.

Pulls structured line items from messy documents (invoices, BOLs, claims, statements) and validates them against schema. Confidence below threshold escalates to a reviewer with the agent's draft pre-filled.

Validation & Match Agent.

Cross-checks extracted values against the matching source record (PO, ASN, claim file, ledger entry). Flags mismatches as drift records with the specific field and the source values cited.

Reconciliation Agent.

Drafts the write-back. Auto-applies only within tenant-set thresholds (low value, clean match, reversible). Anything ambiguous waits for a human.

Trust architecture

Citation or strip.

Every extracted value cites a specific page and bounding box. Every reconciliation cites the source record it matched against. Uncited output is discarded before it reaches a human.

Fail closed, never fabricate.

If OCR is unavailable, a labeled deterministic fallback keeps the pipeline green rather than failing the document. If the validator cannot resolve, the agent escalates. Never invents a match.

Audit-logged, every action.

Every capture, extraction, validation, and write-back lands in the same tamper-evident audit log a human action does. The agent cannot forge or skip the audit.

Human-approved on the write.

Read tools run automatically. Write tools (post a payable, update a ledger, close a record) are dedicated, gated, and require either tenant policy approval or a human token.

What ships first

Phase 1 ships capture and extraction with a per-field confidence score. Validation and reconciliation graduate per tenant once measured agree-rate clears the threshold. We do not ship auto-write as a default. Each tenant configures their own confidence band and reversibility policy.

The thesis

A document is just structured data waiting for someone to type it in.

The cost of that typing, in time, in errors, in re-keys, in delayed downstream work, is the cost most operations teams have stopped measuring because it just lives inside everything. We measure it, then we replace it.

How the engine works

01
Ingest

Any format the work arrives in.

PDFs, scans, photos, emails, spreadsheets, EDI, faxes, the document type your team complains about every Monday. We meet documents where they land, before they become a queue.

02
Classify

Know what it is before deciding what to do with it.

Document-type detection, sub-type routing, priority and urgency signals. The classifier is the first thing we tune for your domain, because every downstream decision rides on it.

03
Extract

Fields, tables, signatures, line items, every cell.

Layout-aware extraction across forms, contracts, and unstructured text. Tables come out as tables, signatures as signatures, dates as ISO. The output is what a downstream system can consume, not a transcript.

04
Verify

Confidence per field, validated against the rules that matter.

Per-field confidence, cross-field consistency, lookups against your master data, and policy checks before anything posts. The 2% the model is unsure about goes to a human queue, with reasoning attached.

05
Route

Into the system that needs it, with the audit it requires.

ERPs, CRMs, claims systems, ticketing, mainframes, custom internal APIs. Every routed payload is timestamped, hash-sealed, and traceable to the source page it came from.

A glimpse of the work

aelix.app · documents · invoice_38274.pdf
InvoiceINV-382742026-05-14ACME Manufacturing Co.12 Industrial Park Rd · Detroit, MI 48201BILL TOAelix EchoDESCRIPTIONQTYUNITAMOUNTSteel brackets, 6mm200$8.40$1,680.00Aluminum spacers, 12mm80$3.20$256.00Hex bolts, M8x401200$0.85$1,020.00Nylon washers, M81200$0.12$144.00Steel plate, 18ga30$42.00$1,260.00Bearing assembly, 6204-2RS24$18.50$444.00Misc fasteners, assorted1$80.00$80.00Subtotal$14,820.00Tax (7%)$1,037.40Total$15,857.40Net 30 · ACME Manufacturing Co. · Thank you for your business.

Extracted

Streaming
schema · invoice_v3 · layout-aware
doc_typeinvoice
99%
invoice_noINV-38274
99%
vendorACME Manufacturing Co.
98%
date2026-05-14
99%
bill_toAelix Echo
97%
line_items7 rows
96%
subtotal$14,820.00
99%
tax$1,037.40
99%
total$15,857.40
99%
1.4s · 14 fields · sealed
Audit hash 7e23...

Live extraction trace · invoice

Outcomes

80%

Reduction in document handling time

99.5%+

Field-level extraction accuracy

Days → mins

Typical document turnaround

Measured across Aelix Echo pilot deployments, 2025-2026

Built for every sector

Seven industries, the documents that run them, and the time we give back.

01
LegalTech
Contracts, case law, redactions
75% time saved
02
FinTech
Statements, KYC, invoices
10x faster processing
03
Healthcare
Claims, EHR, prior authorization
82% error reduction
04
Manufacturing
BOMs, MSDS, work orders
98% BOM accuracy
05
Logistics
Bills of lading, customs, PODs
85% touchless
06
Energy & Utilities
Filings, inspections, permits
4x filing turnaround
07
Real Estate & Construction
Leases, drawings, building permits
80% faster abstracts

How we work

A short discovery, a measured rollout, and a long, quiet operate.

01

Discover

We sit with the team that processes the documents today, watch the work, and find where time and accuracy quietly leak.

02

Train

Domain models fine-tuned on your document mix. Layouts, vocabulary, and edge cases captured before we go anywhere near production.

03

Deploy

Behind a feature flag, with confidence thresholds set conservatively. We watch the first thousand documents together, not over email.

04

Operate

Exception queues, model recalibration, regression tests against real traces. The system gets quietly better while the team gets visibly faster.

A note from the practice

The model isn't done when it's accurate. It's done when the field it's unsure about gets the right human, with the right context, before the downstream system needs an answer. That's the part we work on after the easy 98%.

The Aelix Echo document practice

When a person shouldn't be typing it in, we're here.

45-minute working session · One real document type, extracted live