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 2025. All Rights Reserved by Andrey Deryabin

If you want to launch big vessels, go where the water is deep.

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 2025. All Rights Reserved by Andrey Deryabin

If you want to launch big vessels, go where the water is deep.

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 2025

All Rights Reserved by Andrey Deryabin

If you want to launch big vessels,

go where the water is deep.