Veltora

Pitch 2026

Veltora

Pitch 2026

15 - 20

minutes.

Every time a customer asks about a part.

Every day. In every parts business in Austria.

So funktioniert es heute.

Der Kunde stellt eine Anfrage – telefonisch oder über Messaging-Dienste.

UNSTRUKTURIERTE EINGABE

Ein Mitarbeiter erfasst die Anfrage manuell und sichert Informationen (Screenshots, Notizen, Chatverlauf).

MANUELLE ERFASSUNG

Fahrzeugdaten und Anforderungen werden aus Dokumenten extrahiert und händisch interpretiert.

FEHLERANFÄLLIGE INTERPRETATION

Das passende Ersatzteil wird über einzelne, voneinander getrennte Katalogsysteme identifiziert.

SEPARATE SYSTEME

Preise werden über mehrere Quellen hinweg manuell recherchiert und verglichen.

MEHRFACHRECHERCHE

Verfügbarkeiten werden einzeln bei unterschiedlichen Lieferanten geprüft.

KEINE ZENTRALE ÜBERSICHT

Für spezifische Produkte sind zusätzliche Rechercheschritte in weiteren Systemen notwendig.

SYSTEMBRÜCHE

Alle Informationen werden manuell in ein Angebot übertragen (Dokument, Chat oder E-Mail).

MEDIENBRUCH

Preise und Margen werden individuell kalkuliert.

MANUELLE KALKULATION

Das Angebot wird manuell erstellt und formatiert.

ZEITAUFWAND

Der Versand erfolgt über Messaging oder E-Mail – ohne Systemintegration.

KEIN WORKFLOW

Keine strukturierte Nachverfolgung: Anfragen bleiben unbeantwortet, gehen verloren oder werden nicht weiterverarbeitet.

KEIN FOLLOW-UP

Twelve Steps

Fifteen Minutes minimum.

Fifteen Minutes
minimum.

Per request.

At €35/h × 50h/month = €1,750 lost per workshop, per month.

What if the operator only

had to make one decision?

No searching.

No calculating.

No formatting.

No remembering.

Just: this part — yes or no?

If it's so obvious, why hasn't it been built?

If it's so obvious, why
hasn't it been built?

Vision-language models matured in 2024.

Identifying parts from a photo is now commercially viable, not a research demo.

B2B catalogs opened up.

TecDoc, Partslink24 and OEM finders expose APIs the engine can talk to directly.

WhatsApp Business + Make/n8n.

Distribution and orchestration without building heavy backend, channel infra is solved.

VELTORA

Automotive Intelligence Platform.

Request → Offer → Order. In under 2 minutes.

Der Kunde fragt an

UNSTRUKTURIERTE EINGABE

Der Kunde fragt an

UNSTRUKTURIERTE EINGABE

Three types of requests.
One system.

TYPE A

Standard inquiry (no part number).

Customer describes the part without knowing the reference number.

Without Veltora

15–20 min

3–5 min

TYPE B

Known part number.

Customer already knows the OE or article number direct ordering flow.

Without Veltora

5–10 min

1–2 min

TYPE C

Bulk / multi-vehicle (garage B2B).

A workshop submits requests for 3–10 vehicles simultaneously. Structured, and processed as a batch.

Without Veltora

30–90 min

5–10 min

Every channel.
Every input.
One pipeline.

Customers write where they already write. Veltora normalizes the chaos before it ever reaches an operator's screen.

0.1

Photo of the broken part. Voice note. Done.

Photo of the broken part.

Voice note. Done.

Photo of the broken part.

Voice note. Done.

0.2

B2B fleets. Bot-style requests with VIN.

B2B fleets. Bot-style

requests with VIN.

B2B fleets. Bot-style

requests with VIN.

0.3

Web chat

Public site. Drag a Kfz-Schein, hit send.

Public site. Drag a

Kfz-Schein, hit send.

Public site. Drag a

Kfz-Schein, hit send.

0.4

Forward the supplier's PDF. We parse it.

Forward the supplier's PDF.

We parse it.

Forward the supplier's PDF.

We parse it.

0.5

SMS

Older shops. Plain text. Same pipeline.

Older shops. Plain text.

Same pipeline.

Older shops. Plain text.

Same pipeline.

ACCEPTED INPUTS

Text

Photo

Document

VIN Code

Parts

Invoice

WHAT HAPPENS NEXT

Advanced AI processing extracts all relevant information automatically. VIN numbers, brand details, model year, engine specs, and part names are identified and structured before human review.

ACCEPTED INPUTS

Text

Photo

Document

VIN Code

Parts

Invoice

WHAT HAPPENS NEXT

Advanced AI processing extracts all relevant information automatically. VIN numbers, brand details, model year, engine specs, and part names are identified and structured before human review.

One inbox. One pipeline. Zero copy-paste.

Fluent in every
catalog that matters.

Not a wrapper around one database. A router across the catalogs the entire aftermarket already runs on, chosen per request, in milliseconds.

Not a wrapper around one database. A router across the catalogs the entire aftermarket already runs on,

chosen per request, in milliseconds.

TECDOC

Aftermarket cross-reference

PARTSLINK24

VIN → OE parts

PARTSLINK24

VIN → OE parts

LIQUI MOLY

Oil finder

LIQUI MOLY

Oil finder

OSRAM

Bulb & lighting compatibility

BOSCH

Wipers & ignition

BOSCH

Wipers & ignition

OE/AFTERMARKET

Fallback supplier graph

OE/AFTERMARKET

Fallback supplier graph

The numbers
dont lie

Figures based on 8 executions per day

Saved every month

0€

Total monthly cost

Return on infrastructure cost.

Forgotten requests.

What happens
after the offer.

Most platforms stop at the quote. The expensive moments live afterwards, in the silence, the second thoughts, the parts that get forgotten.

branch 01

Customer orders

Confirmed

Order written to ledger. Supplier triggered. Lieferschein and Rechnung queued automatically.

→ Fulfillment

branch 02

Customer modifies

Re-run.

Pipeline replays with the updated input. New offer, same conversation, in seconds — not the next morning.

→ Updated quote

branch 03

Customer goes silent

Followed up.

+24h auto follow-up on the same channel. +72h second nudge. Nothing falls through. Nothing forgotten.

→ 0 lost requests

A request never dies. Every conversation lives in the ledger, searchable,

re-quotable, follow-uppable by a system that doesn't forget on a Friday afternoon.

A request never dies. Every conversation lives in the ledger, searchable,

re-quotable, follow-uppable by a system that doesn't forget on a Friday afternoon.

€

From confirmation to
the customer's hands.

From confirmation to the customer's hands.

The offer is the easy part. Veltora's value compounds through every step that comes after — the steps competitors leave on a clipboard.

Order placed

Wholesaler

Order written to ledger. Supplier triggered. Lieferschein and Rechnung queued automatically.

Goods received

QC & repack

Parts arrive. Veltora checks them in, flags damage, repacks where needed, all logged against the original request.

Documents

Lieferschein · Rechnung

Delivery note and invoice generated automatically with sequential numbering and the customer's commercial terms.

Handover

Delivery or pickup

Customer chooses. Either way, they get a status link — not a phone call asking if it's ready.

LIEFERSCHEIN

VLT-2826

Rechnung

VLT-9860

Generated automatically

Sequential numbering. Austrian VAT. Customer terms.

Every document a tax inspector would expect — generated the second the order clears, attached to the same conversation thread the customer started in.

Three ways Veltora earns.

Most platforms stop at the quote. The expensive moments live afterwards, in the silence, the second thoughts, the parts that get forgotten.

Today · Vienna

Margin per order

Veltora's first revenue stream is the workshop itself. Every quote becomes an order; every order earns a margin between cost and offer.

Q2 2027 · DACH

SaaS per workshop

Once the engine is proven inside one workshop, it ships as a subscription to the next 6,400 in DACH. Same software, multiplied.

Q2 2027 · DACH

SaaS per workshop

Once the engine is proven inside one workshop, it ships as a subscription to the next 6,400 in DACH. Same software, multiplied.

2028 · Network

Take rate · wholesalers

When the network matures, each transaction routed through Veltora pays a small fee back. The marketplace funds itself.

2028 · Network

Take rate · wholesalers

When the network matures, each transaction routed through Veltora pays a small fee back. The marketplace funds itself.

A request never dies. Every conversation lives in the ledger, searchable,

re-quotable, follow-uppable by a system that doesn't forget on a Friday afternoon.

Every mechanic shop, every garage, every distributor in Austria has the same problem.

Every mechanic shop, every
garage, every distributor in \Austria has the same problem.

Every mechanic shop, every garage, every distributor in Austria has the same problem.

6,400+ workshops · ~12 min lost per quote · €480M annual waste across Austria.

6,400+ workshops · ~12 min lost per quote

€480M annual waste across Austria.

Source: Statistik Austria · WKO Fahrzeughandel 2024

One workshop. One real workflow. One paying customer from day one, not a hypothesis, an operating system.

6,400+

Workshops

~12 min

Lost / quote

€ 480M

Annual waste

Gewerbepark Klosterneuburg: 2-container warehouse.

Workshops

Zone 1

Zone 2

Zone 3

Klosterneuburg / Tulln

Vienna

Lower Austria

A request never dies. Every conversation lives in the ledger, searchable,

re-quotable, follow-uppable by a system that doesn't forget on a Friday afternoon.

Veltora can be the platform
that solves it for all of them.

One workshop today. The whole industry tomorrow.

Same engine, multiplied across a country that's been waiting for it.

One workshop today. The whole industry tomorrow.

Same engine, multiplied across a country that's been waiting for it.

Q3 2026

Vienna workshop · live

Q4 2026

3 pilot workshops

Q2 2027

Public SaaS · DACH

2028

Wholesaler network

The founder

Artur Udartsev

BSc WU Wien · LL.M. · Stanford ML & AI · Klosterneuburg, AT

Sector

Years inside the automotive parts trade, operator-side, not from the outside.

Sector

Years inside the automotive parts trade, operator-side, not from the outside.

Distribution

First paying workshop already secured in Vienna. Day-one revenue, not a hypothesis.

Distribution

First paying workshop already secured in Vienna. Day-one revenue, not a hypothesis.

Building Veltora from inside the workshop, not from a deck.

Building Veltora from inside the

workshop, not from a deck.

"I've seen this flow from the inside.
That's why I know exactly where it breaks."

Background

Sector

Years operating inside trade, supply chain and import coordination — DACH region. BSc International Business (WU Wien), specialisation Logistics & Supply Chain Management.

Knowledge

Operations-first founder. Ran Serex GmbH for 4 years, procurement coordination, supplier networks across EU and CIS, online commerce and logistics. Built the workflow Veltora automates before automating it.

Method

Operator-first. Every feature ships against a real request, not a roadmap.

Method

Operator-first. Every feature ships against a real request, not a roadmap.

What could go wrong
and how we de-risk it.

What could go wrong and how we de-risk it.

Objection 01

What if catalog

APIs change?

What if catalog

APIs change?

Vision-first fallback. The engine reads the part visually when an API breaks, and a manually-curated catalog covers the long tail. No single point of failure.

Objection 03

What if a workshop

won't change behavior?

What if a workshop

won't change behavior?

Zero adoption curve — it's WhatsApp. The operator does what they already do. We replace the back-office, not the front-of-house.

Objection 03

Take rate · wholesalers

18 months of workflow data + integrated wholesalers compound. The moat is what we ship after the first thousand orders, not the first one.

DEMO

Now, see it.

A real request from a real workshop. Watch every step happen, from a WhatsApp photo to a branded offer in the customer's inbox.

0 s

Total runtime

0 stages

End to end

0 photo

Single input

0 tiers

PDF output

veltora · workshop console

Standby·0.0s

Markus Krause

via WhatsApp · online

Servus, I need front brake pads for my Audi. Can you send me an offer?

09:14

Kfz-Schein.jpg · 2.4 MB

vehicle registration

09:14

😊Type a message…●

pipeline · 00/08

CPU 12%·Tokens 0

Message received

GPT-4o Vision — extracting VIN

VIN decoded

Work order generated

Operator: cost price

Margins applied — 3 tiers

PDF offer generated

Delivered — WhatsApp + Email

Old way

15–20 min

Veltora

0.0s

veltora · console

—·0.0s

Markus Krause

WhatsApp · online

Servus, I need front brake pads for my Audi. Can you send me an offer?09:14

Kfz-Schein.jpg

vehicle registration09:14

😊Type a message…●

Standby

00/08

Message received

GPT-4o Vision — VIN extract

VIN decoded

Work order generated

Cost lookup

Margins — 3 tiers

PDF offer rendered

Delivered

Old way

15–20 min

Veltora

0.0s

TRY IT

Now, try it.

Veltora · Live-Demo

Testen Sie den VELTORA-Workflow.

Laden Sie Fotos, Fahrzeugschein oder Teilenummern hoch. Die Anfrage wird im MVP-Backend strukturiert erfasst und bearbeitet.

oder eigene Anfrage eingeben

Kfz-Schein hochladen · VIN einfügen · Teilnummer angeben

Hinweis: Dies ist ein Pilot-Workflow. Die Bearbeitung erfolgt im aktuellen MVP-Stand manuell bzw. semi-manuell im Backend.

Veltora · Live-Demo

Testen Sie den VELTORA-Workflow.

Laden Sie Fotos, Fahrzeugschein oder Teilenummern hoch. Die Anfrage wird im MVP-Backend strukturiert erfasst und bearbeitet.

oder eigene Anfrage eingeben

VIN · Kfz-Schein · Teilnummer

Hinweis: Dies ist ein Pilot-Workflow. Die Bearbeitung erfolgt im aktuellen MVP-Stand manuell bzw. semi-manuell im Backend.

office@veltora.at

VELTORA

Contact: office@veltora.com

15 - 20

15 - 20

minutes.

Every time a customer asks about a part.

Every day. In every parts business in Austria.

Twelve Steps

Twelve Steps

Fifteen Minutes minimum.

Fifteen Minutesminimum.

Per request.

Per request.

What if the operator only

What if the operator only

had to make one decision?

had to make one decision?

Just: this part — yes or no?

If it's so obvious, why hasn't it been built?

If it's so obvious, whyhasn't it been built?

VELTORA

VELTORA

VELTORA

Der Kunde fragt an

UNSTRUKTURIERTE EINGABE

Der Kunde fragt an

UNSTRUKTURIERTE EINGABE

Three types of requests.One system.

Three types of requests.One system.

Every channel.Every input.One pipeline.

Every channel.Every input.One pipeline.

Fluent in everycatalog that matters.

Fluent in everycatalog that matters.

The numbersdont lie

The numbersdont lie

What happensafter the offer.

What happensafter the offer.

From confirmation tothe customer's hands.

From confirmation tothe customer's hands.

From confirmation to the customer's hands.

Three ways Veltora earns.

Three ways Veltora earns.

Three ways Veltora earns.

Every mechanic shop, every garage, every distributor in Austria has the same problem.

Every mechanic shop, everygarage, every distributor in \Austria has the same problem.

Every mechanic shop, every garage, every distributor in Austria has the same problem.

Veltora can be the platformthat solves it for all of them.

Veltora can be the platformthat solves it for all of them.

Veltora can be the platformthat solves it for all of them.

The founder

The founder

The founder

Artur Udartsev

Artur Udartsev

Artur Udartsev

BSc WU Wien · LL.M. · Stanford ML & AI · Klosterneuburg, AT

BSc WU Wien · LL.M. · Stanford ML & AI · Klosterneuburg, AT

BSc WU Wien · LL.M. · Stanford ML & AI · Klosterneuburg, AT

"I've seen this flow from the inside.That's why I know exactly where it breaks."

"I've seen this flow from the inside.That's why I know exactly where it breaks."

"I've seen this flow from the inside.That's why I know exactly where it breaks."

What could go wrongand how we de-risk it.

What could go wrongand how we de-risk it.

What could go wrong and how we de-risk it.

Now, see it.

Now, see it.

Now, try it.

Now, try it.

office@veltora.at

Contact: office@veltora.com

Contact: office@veltora.com

Fifteen Minutes
minimum.

If it's so obvious, why
hasn't it been built?

Three types of requests.
One system.

Three types of requests.
One system.

Every channel.
Every input.
One pipeline.

Every channel.
Every input.
One pipeline.

Fluent in every
catalog that matters.

Fluent in every
catalog that matters.

The numbers
dont lie

The numbers
dont lie

What happens
after the offer.

What happens
after the offer.

From confirmation to
the customer's hands.

From confirmation to
the customer's hands.

Every mechanic shop, every
garage, every distributor in \Austria has the same problem.

Veltora can be the platform
that solves it for all of them.

Veltora can be the platform
that solves it for all of them.

Veltora can be the platform
that solves it for all of them.

"I've seen this flow from the inside.
That's why I know exactly where it breaks."

"I've seen this flow from the inside.
That's why I know exactly where it breaks."

"I've seen this flow from the inside.
That's why I know exactly where it breaks."

What could go wrong
and how we de-risk it.

What could go wrong
and how we de-risk it.