Turn freight documents into actionable intelligence.
Novelavia uses artificial intelligence to automate freight document processing, reduce manual data entry, and simplify logistics operations.
Intelligence behind every shipment. Built for carriers, brokers, dispatchers and freight accounting teams in the United States.
The problem
Freight runs on documents. Most of them are still retyped by hand.
Dispatchers, brokers and accounting teams move information from emails and PDFs into the systems that run the business, one field at a time. It is slow, error-prone and hard to scale as load volume grows.
- Reviewing freight emails and opening every PDF attachment
- Re-keying pickup and delivery details into the TMS
- Checking agreed rates, accessorials and payment terms
- Comparing documents from different stages of a load
- Chasing missing signatures and proofs of delivery
- Preparing invoices and instructions for drivers
Where shipment information lives today
- Email inboxes
- PDF attachments
- Scanned paperwork
- Spreadsheets
- TMS and accounting tools
One structured, validated shipment record
What Novelavia is built to produce from every document.
How Novelavia works
From inbox to structured data in four steps
A pipeline designed around how freight teams already work: documents in, verified data out, with a person in the loop wherever the AI is not sure.
- 01
Upload
Import logistics documents from files today, and from email and other integrations in the future.
rate-con-NV-10482.pdfUploadedinvoice-ART-30418.pdfQueued - 02
Understand
AI identifies the document type and extracts the information that matters for the load.
Rate ConfirmationLoad # NV-2026-104822 stops$2,450.00Net 30 - 03
Validate
Validation highlights inconsistencies, missing information and uncertain fields for a person to review.
- Rate line items add up
- Stop dates in order
- Carrier signature missing
- 04
Automate
Verified data becomes structured information ready for operational workflows and system integrations.
JSONCSVTMS · plannedAccounting · planned
Illustrations use values from the fictional sample documents in the demo.
Supported document types
Built for the documents that move freight
The MVP starts with rate confirmations, the document every load begins with. Other document types follow as the platform grows.
- MVP focus
Rate Confirmation RC
Agreed rate, stops, appointments, equipment, accessorial terms and broker requirements.
Try it in the demo - In development
Bill of Lading BOL
Shipper and consignee, piece count, weight, commodity and seal numbers.
- In development
Proof of Delivery POD
Delivery date and time, receiver signature and exceptions such as shortages or damage.
- In development
Freight Invoice INV
Line items, accessorials and totals, compared with the agreed rate.
- Planned
Carrier Agreement CA
Parties, payment terms, insurance requirements and key obligations.
- Planned
Lumper Receipt LR
Loading or unloading fees, facility, date and payment method for reimbursement.
- Planned
Fuel Receipt FR
Gallons, price, location and date for fuel reconciliation.
Product capabilities
An automation layer for freight documents
Five capabilities, each clearly labeled with where it stands today. We would rather show you a working proof of concept than promise a finished platform.
AI Rate Confirmation Processing
Designed to read rate confirmations and return structured shipment data, with a confidence score for every field.
Open the demo// sample output from a fictional rate confirmation
load_number: "NV-2026-10482"
broker.mc: "MC-9917735"
carrier.usdot: "9921846"
pickup.date: "2026-11-17"
delivery.city: "Dallas, TX"
equipment: "Dry Van 53 ft"
weight_lbs: 42000
rate.total: 2450.00
payment_terms: "Net 30"
detention: "$50/hr after 2 hrs"
Intelligent Document Classification
Recognizes what each document is before extracting anything, so rate confirmations, BOLs, PODs and invoices each get the right treatment.
AI Document Validation
Business rules check every result before anyone relies on it. Uncertain or inconsistent values are routed to a person, and missing values are reported instead of guessed.
Automated Dispatch Preparation
Turns verified load data into clear driver instructions: stops, appointment times, equipment, reference numbers and special requirements.
Document-to-Workflow Automation
Verified data is exported as JSON or CSV today. Direct delivery to the systems your team already uses is planned, so nobody has to type a load in twice.
Interactive product preview
Every value, traced back to the page
Hover a field to see exactly where it came from. This preview uses the fictional sample rate confirmation and its predefined results.
Novelavia Document Intelligence
Sample data · predefined results
Classified as
Rate Confirmation 99%
Business benefits
Less paperwork. More attention on the freight.
Novelavia is designed around outcomes freight operators care about. We will publish measured results only after they have been verified with pilot customers.
- Less manual data entry
- Capture load details once from the source document instead of retyping them into every system.
- Fewer costly mistakes
- Validation flags missing fields, conflicting dates and rate mismatches before they reach dispatch or billing.
- Cleaner billing
- Designed to compare invoices with rate confirmations and catch missing proof of delivery before payment is delayed.
- Faster dispatch preparation
- Planned: driver-ready instructions built from the same verified data, without copying and pasting.
- Room to grow
- Designed to help teams handle more loads without adding administrative work at the same pace.
- People stay in control
- Uncertain values go to a reviewer, every edit is tracked, and nothing is silently invented.
Technology
AI-native, designed for Google Cloud
Multimodal Gemini models read documents in context. Deterministic validation checks every result. The planned architecture is lean and serverless on Google Cloud, designed to keep documents private and scale with volume.
Gemini on Gemini Enterprise Agent Platform
Prototype
Cloud Run
MVP
Cloud Storage
MVP
Firestore
MVP
Secret Manager
MVP
Identity Platform
MVP
- 1UploadCloud Storage
- 2ProcessCloud Run
- 3UnderstandGemini · Document AI
- 4ValidateBusiness rules
- 5ReviewHuman in the loop
- 6DeliverAPI · JSON · CSV
Proposed production pipeline. Most services are planned, not yet in production.
Help shape Novelavia from day one.
We are looking for U.S. carriers, brokers, dispatch services and freight accounting teams to pilot the MVP. Early access members try new capabilities first and directly influence the roadmap.