Social Ai Agent

AI social commerce dashboard for managing conversations, leads, products, media uploads, platform connections, and subscription billing in one workspace.

Comment

Inspiration

Social Ai Agent was inspired by small businesses that depend on social media conversations to make sales, but often do not have enough time or staff to manage every message, comment, product question, lead, and follow-up across different platforms.

Many sellers receive customer interest through WhatsApp, TikTok, YouTube, Facebook, Instagram, and other channels, but their conversations, products, media, and billing workflows are scattered. Social Ai Agent brings those pieces into one workspace.

What it does

Social Ai Agent is an AI-assisted social commerce dashboard for managing customer conversations, leads, products, media, platform connections, and subscription billing.

The app includes a Streamlit dashboard, a FastAPI backend, AI-assisted replies, product memory, client tracking, media uploads, multi-platform conversation views, OAuth connection flows, admin tools, webhooks, and monthly subscription billing with PayPal, Paystack, and Dodo Payments.

How I built it

I built Social Ai Agent with Python, Streamlit, FastAPI, SQLAlchemy, and a database layer that supports PostgreSQL in production and SQLite locally.

The dashboard is built in Streamlit and provides pages for the main workspace, client onboarding, media management, product memory, live chat, analytics, platform connections, settings, billing, admin views, and documentation.

The backend exposes REST API endpoints for authentication, products, clients, media, conversations, billing, platform connections, uploads, webhooks, and admin activity. The AI reply service uses the Google Gemini API when configured and includes a fallback reply system so the app can still respond when the AI provider is unavailable.

Challenges I ran into

One challenge was combining many parts of a SaaS product into one working system: authentication, AI replies, product memory, media uploads, customer conversations, platform integrations, billing, and admin monitoring.

Another challenge was making the app usable even when some external services are not configured yet. To handle this, I added fallback responses, demo-friendly data, deployment checks, and readiness indicators for AI, billing, database, and platform setup.

Payment integration was also challenging because PayPal, Paystack, and Dodo Payments each handle checkout sessions, approval links, callbacks, and status updates differently.

Accomplishments that I'm proud of

I am proud that Social Ai Agent is more than a simple chatbot. It is a working social commerce command center with a dashboard, backend API, AI-assisted replies, product and media storage, conversation management, billing flows, webhooks, and admin tools.

I am also proud of building the app in a way that can run locally with SQLite but also support production deployment with PostgreSQL, Render, and Streamlit Community Cloud.

What I learned

I learned how to structure a full-stack Python app with a Streamlit frontend and FastAPI backend. I also learned more about AI-assisted workflows, OAuth platform connections, payment provider integration, database-backed product memory, and building dashboards around real business operations.

What's next for Social Ai Agent

Next, I want to improve the live platform integrations, add deeper automation for follow-ups, expand analytics, improve onboarding, and make the AI assistant more action-oriented across conversations, products, customer records, and billing workflows.

Partnership and acquisition interest

Social Ai Agent is open to partnership, licensing, sponsorship, or acquisition conversations. If you are interested in using, funding, distributing, or acquiring the project, contact me through my Devpost profile or email me at: saintvellian@gmail.com

Built With

Share this project:

Updates

posted an update —

Social Ai Agent is available for partnership, licensing, sponsorship, or acquisition discussions.

The project is an AI-assisted social commerce dashboard for small businesses, with product memory, lead tracking, AI reply suggestions, platform connection flows, media uploads, admin tools, and subscription billing.

If you are interested in collaborating, investing, licensing, or acquiring the app, contact me through Devpost or by email:saintvellian@gmail.com

Log in or sign up for Devpost to join the conversation.

posted an update —

Social Ai Agent is now live on Devpost.

I built it as an AI-assisted social commerce dashboard for small businesses to manage conversations, leads, products, media uploads, platform connections, and subscription billing from one workspace.

Current features include a Streamlit dashboard, FastAPI backend, Gemini-powered reply suggestions, product memory, client tracking, multi-platform conversation views, admin tools, webhooks, and billing flows with PayPal, Paystack, and Dodo Payments.

Log in or sign up for Devpost to join the conversation.