Uploaded October 2026 | Updated November 2026, 5 days ago
A single internal app now handles Railway's marketing content, email, social posts, media, creator relationships and launch planning. Mahmoud Abdelwahab built it with coding agents, but getting the features to run was only the beginning. The difference between a quick SaaS replacement and a tool his team wants to open every day lives in the details: background saving, responsive interactions, reversible actions and consistent screens. His starting point is a structured codebase with reusable abstractions, so each new feature follows existing patterns instead of inventing another way to do the same work.
Abdelwahab works through the product behaviors he has learned to request explicitly. Autosave should preserve progress without blocking typing; optimistic updates should make suitable actions feel immediate and recover when the operation fails. Undo and archiving give people room to change their minds. URL state makes a filtered view shareable, while infinite scrolling removes repetitive navigation. Referencing an existing page helps the agent reuse the app's interface and data fetching conventions. He then packages recurring design and deployment preferences into agent skills. The final step is a verification loop that opens the running application and exercises the feature from start to finish, because a successful build and passing tests can still leave a broken experience. These are practical lessons from an internal tool, without a claim of measured financial savings.
Speaker info:
- linkedin.com/in/thisismahmoud
- github.com/m-abdelwahab
Timestamps:
00:00 - Replacing Railway's internal marketing tools
2:32 - Structure and abstractions for consistent features
4:03 - Product polish agents need to be asked for
4:57 - Autosave and collaboration
6:16 - Optimistic updates, undo and archiving
8:45 - URL state and navigation
9:38 - Reference pages and agent skills
11:47 - Verify the running feature from start to finish
---
Watch along with the corrected transcript, summary, timestamps and resources on this talk's official AIE page:
https://ai.engineer/talks/oixzVKMAXG0-youtube-6fcfb5ae5a2bdc32a5b2
Join our upcoming AI Engineer conferences (see https://ai.engineer landing page):
- AIEi Shanghai, November 5–6 — the first for meeting top Chinese labs and engineers
- AIE CODE in SF, November 10–12 — the top AI Coding conference returns
- AIEi Sydney, December 7–8 — the first AIEi run alongside NeurIPS 2026 in Sydney
---
A single internal app now handles Railway's marketing content, email, social posts, media, creator relationships and launch planning. Mahmoud Abdelwahab built it with coding agents, but getting the features to run was only the beginning. The difference between a quick SaaS replacement and a tool his team wants to open every day lives in the details: background saving, responsive interactions, reversible actions and consistent screens. His starting point is a structured codebase with reusable abstractions, so each new feature follows existing patterns instead of inventing another way to do the same work.
Abdelwahab works through the product behaviors he has learned to request explicitly. Autosave should preserve progress without blocking typing; optimistic updates should make suitable actions feel immediate and recover when the operation fails. Undo and archiving give people room to change their minds. URL state makes a filtered view shareable, while infinite scrolling removes repetitive navigation. Referencing an existing page helps the agent reuse the app's interface and data fetching conventions. He then packages recurring design and deployment preferences into agent skills. The final step is a verification loop that opens the running application and exercises the feature from start to finish, because a successful build and passing tests can still leave a broken experience. These are practical lessons from an internal tool, without a claim of measured financial savings.
Speaker info:
- linkedin.com/in/thisismahmoud
- github.com/m-abdelwahab
Timestamps:
00:00 - Replacing Railway's internal marketing tools
2:32 - Structure and abstractions for consistent features
4:03 - Product polish agents need to be asked for
4:57 - Autosave and collaboration
6:16 - Optimistic updates, undo and archiving
8:45 - URL state and navigation
9:38 - Reference pages and agent skills
11:47 - Verify the running feature from start to finish
---
Watch along with the corrected transcript, summary, timestamps and resources on this talk's official AIE page:
https://ai.engineer/talks/oixzVKMAXG0-youtube-6fcfb5ae5a2bdc32a5b2
Join our upcoming AI Engineer conferences (see https://ai.engineer landing page):
- AIEi Shanghai, November 5–6 — the first for meeting top Chinese labs and engineers
- AIE CODE in SF, November 10–12 — the top AI Coding conference returns
- AIEi Sydney, December 7–8 — the first AIEi run alongside NeurIPS 2026 in Sydney
---










