While thinking about how people learn new skills, I noticed something interesting. There are thousands of free tutorials online and hundreds of paid courses, yet many people still struggle to learn. The biggest issue isn't always access to content—it's having someone to guide you, answer questions, and keep you accountable.
At the same time, almost everyone knows something that someone else wants to learn.
That made me wonder: instead of paying for courses, what if people could exchange knowledge directly?
That's how TradeSkill was born.
*What it does
*
TradeSkill is a peer-to-peer skill exchange platform where users teach one skill in return for learning another.
For example, if I know Photoshop and want to learn Python, and someone else knows Python but wants to learn Photoshop, TradeSkill matches us. We schedule a session, meet over a video call, and teach each other.
No money is involved. Skills become the currency.
The platform also includes chat, live video sessions, ratings, and a matching system that helps users find suitable learning partners.
*Why I think it's different
*
Most learning platforms focus on delivering content from one instructor to many learners.
TradeSkill focuses on connecting people with complementary skills.
Unlike recorded courses, every session is interactive. Learners can ask questions, receive feedback immediately, and learn at their own pace.
Unlike YouTube, there is accountability because another person is investing their time too.
Most importantly, every user contributes something to the community instead of only consuming content.
*How I built it
*
The frontend is built with React and deployed on Vercel.
For the backend, I used AWS DynamoDB to store user profiles, conversations, matches, ratings, and session information.
*The application currently uses six DynamoDB tables:
*
- Users
- Matches
- Messages
- InboxHeads
- VideoSessions
- Ratings
For live learning sessions, I integrated Jitsi Meet so users can join video calls directly from the platform without requiring additional software.
The matching system compares the skills users can teach with the skills they want to learn and generates compatibility scores to recommend potential partners.
*Challenges I ran into
*
Designing the database was more challenging than I expected because the application isn't just another chat platform.
I needed separate data structures for user profiles, conversations, inbox previews, video sessions, ratings, and skill matches while keeping reads and writes efficient in DynamoDB.
Another challenge was creating a matching flow that actually feels useful instead of simply listing users with similar interests.
Integrating real-time communication while keeping the overall experience simple also required several iterations.
*What I learned
*
This project gave me hands-on experience with DynamoDB data modeling, integrating third-party video services, and designing applications around user interactions instead of traditional CRUD operations.
I also learned how important product thinking is. Building features is only one part of the process; designing an experience that people would actually use is much harder.
*What's next
*
If I continue developing TradeSkill, I want to add:
- AI-assisted skill matching
- Skill verification through assessments
- Group learning sessions
- Learning streaks and achievements
- Calendar integration
- Recommendation system based on previous sessions
- Mobile application support
The goal is to build a platform where anyone can learn something new without cost by sharing what they already know.