FreightTech Startup

Personal ChatGPT for Logistics

We reframed Tiliport as a private, logistics-specific ChatGPT – but based on each user’s particular logistics data.

Year :

2025

Niche :

B2C Software on B2B Market

Company :

Tiliport - 4th Version

Duration :

1 month

Problem :

ChatGPT can’t help with your real logistics. It doesn’t know your customers, providers, shipments, or deadlines. Freight professionals need answers and reminders grounded in their own data – not general knowledge.

Solution :

Tiliport worked like a teammate. It read your logistics emails (when CC’ed), then answered questions like:

– What’s arriving next week?

– Who didn’t send docs?

– Which quotes weren’t answered?

Plus, it emailed you daily reminders based on what matters to you.

Challenge :

Despite all benefits, users still didn’t CC emails. We realized that users didn’t have this pain - they simply don’t have anything to ask logistics AI in chat. We also realized that only c-level executives can approve tools that access internal data. Employees couldn’t adopt it on their own. Corporate ownership of data and fear of being “watched” overrode the utility.

Summary :

– Built a private, freight-specific ChatGPT with real-time email context

– Learned that personal adoption isn’t possible in an enterprise data environment

– Pivoted to a value proposition teams could adopt without resistance

More Projects

© Copyright 2026. All Rights Reserved by Andrey Deryabin

Structural advantage is operational long before it becomes financial.

FreightTech Startup

Personal ChatGPT for Logistics

We reframed Tiliport as a private, logistics-specific ChatGPT – but based on each user’s particular logistics data.

Year :

2025

Niche :

B2C Software on B2B Market

Company :

Tiliport - 4th Version

Duration :

1 month

Problem :

ChatGPT can’t help with your real logistics. It doesn’t know your customers, providers, shipments, or deadlines. Freight professionals need answers and reminders grounded in their own data – not general knowledge.

Solution :

Tiliport worked like a teammate. It read your logistics emails (when CC’ed), then answered questions like:

– What’s arriving next week?

– Who didn’t send docs?

– Which quotes weren’t answered?

Plus, it emailed you daily reminders based on what matters to you.

Challenge :

Despite all benefits, users still didn’t CC emails. We realized that users didn’t have this pain - they simply don’t have anything to ask logistics AI in chat. We also realized that only c-level executives can approve tools that access internal data. Employees couldn’t adopt it on their own. Corporate ownership of data and fear of being “watched” overrode the utility.

Summary :

– Built a private, freight-specific ChatGPT with real-time email context

– Learned that personal adoption isn’t possible in an enterprise data environment

– Pivoted to a value proposition teams could adopt without resistance

More Projects

© Copyright 2026. All Rights Reserved by Andrey Deryabin

Structural advantage is operational long before it becomes financial.

FreightTech Startup

Personal ChatGPT for Logistics

We reframed Tiliport as a private, logistics-specific ChatGPT – but based on each user’s particular logistics data.

Year :

2025

Niche :

B2C Software on B2B Market

Company :

Tiliport - 4th Version

Duration :

1 month

Problem :

ChatGPT can’t help with your real logistics. It doesn’t know your customers, providers, shipments, or deadlines. Freight professionals need answers and reminders grounded in their own data – not general knowledge.

Solution :

Tiliport worked like a teammate. It read your logistics emails (when CC’ed), then answered questions like:

– What’s arriving next week?

– Who didn’t send docs?

– Which quotes weren’t answered?

Plus, it emailed you daily reminders based on what matters to you.

Challenge :

Despite all benefits, users still didn’t CC emails. We realized that users didn’t have this pain - they simply don’t have anything to ask logistics AI in chat. We also realized that only c-level executives can approve tools that access internal data. Employees couldn’t adopt it on their own. Corporate ownership of data and fear of being “watched” overrode the utility.

Summary :

– Built a private, freight-specific ChatGPT with real-time email context

– Learned that personal adoption isn’t possible in an enterprise data environment

– Pivoted to a value proposition teams could adopt without resistance

More Projects

© Copyright 2026

All Rights Reserved by Andrey Deryabin

Structural advantage is operational

long before it becomes financial.