Automate-LinkedIn is an AI-powered application built with Next.js
that generates LinkedIn posts based on user descriptions. By leveraging Gemini
, the app enables users to easily craft posts, making it perfect for professionals looking to save time and generate creative content for their LinkedIn profiles.
Demo
Here is the demo video :
Here are the screenshots:
Codebase :
Code is accessible through this repository:
Automate-LinkedIn: Your Superhuman LinkedIn Post Writer Expert
Automate-LinkedIn is an AI-powered application built with Next.js that generates LinkedIn posts based on user descriptions. By leveraging Gemini's API, the app enables users to easily craft posts, making it perfect for professionals looking to save time and generate creative content for their LinkedIn profiles.
Video Link: Watch Demo
Features
- AI Post Generation: Users provide a description, and the app generates a professional LinkedIn post based on the input
- Gemini API Integration: Utilizes Gemini's API to process natural language descriptions
- Post Editing: Edit generated posts before finalizing
- Copy to Clipboard: Easy one-click copying of generated posts
- Regenerate Option: Generate alternative versions of your posts
- Responsive Design: Works seamlessly across different devices
- Modern UI: Clean interface with Tailwind CSS styling
Tech Stack
- Next.js 15.1.5
- React 19
- TypeScript
- Tailwind CSS
- Jest for testing
- Google Generative AI (Gemini)
Github Copilot usage
I am using this for the first time so I installed and done the setup :
- Using Inline chat functionality: I found this very useful as it focus only on that snippet on which we want to make changes.
- Usage of Chat Sidebar of github copilot: I also found this very useful as for queries we don't have to go to the browser and find solutions.
- Usage of Just Code ( Ghost Text ) : This is interesting one and it's not that accurate but it knows what will be the upcomingfunctionality:
- Usage of Comments Completion : I found this useful and dangerous as well. It also sometimes reveals id like in below screenshot it provide Google gemini api key ( which doesn't work obviously)
Cases where I struggle :
If we try to generate whole functionality then it doesn't work that's why it's called copilot but still have some decent accuracy.
Works in terminal as well:
Best thing about copilot ( according to me) :
- It setup the jest in my
Nextjs
project and yes I completely shocked by it's accuracy.
- It also helps a lot in
Readme.md
file content :
Summary :
I have used below functionalities of Github copilot and it takes around 5-6 hours
in one go.
-
Just code ( Ghost text) : for completing
ts
functions and unit tests writing. - Writing comments : for adding comments for functions, api's and unit tests.
-
Side Panel (Chat Sidebar) : this is really really useful. It setup my whole
jest
configurations, helps inReadme.md
file, my own queries regarding bug andnextjs
issues solving. - Inline chat: this is very helpful to make changes to small snippet of code.
- Prompt based code: Also like chatgpt I explain my things to copilot and it respond back accordingly.
-
Asking queries according to
@workspace
and#file.tsx
: This is very unique instead of passing the file code we can directly chat with that file and workspace.
Top comments (0)