Inspiration
Every time we launch something, it goes unnoticed. But our random social media posts - news or not - bring traffic.
As an engineering-only team, marketing is a constant pain. We can’t justify a full-time hire, and part-time help keeps losing context, needing our input just to get back up to speed.
We tried existing social media tools, but they only centralize tasks. We’re still left figuring out what to post, when to post, and how to build a real strategy. We end up juggling tools, channels, and creative work on our own.
We knew we weren’t alone! Most teams struggle to market consistently without a full stack of specialists. Even those with dedicated marketing staff face the same chaos, only at scale.
So this weekend, we built what we always wanted: an autonomous marketing agent that works like a full-time teammate, always on, never losing context, and contributing across the entire marketing funnel, from inbound to outbound.
What it does
MAIA is an autonomous marketing agent that monitors your brand’s landscape, collaborates with your team, and transforms business goals into high-impact campaigns. With future integrations, MAIA will be able to execute the strategy seamlessly across channels too.
How we built it
We developed MAIA using large language models (to be migrated to agentic approach) integrated with React/Next, Node.js(for prototyping), and Go services (to serve fast endpoints). Features include onboarding, goal intake, campaign strategy, and content generation, real-time editing, and scheduling.
Challenges we ran into
- Anticipated blocker: Integration with social media platforms is going to take some time get approved.
Accomplishments that we're proud of
- Built the agent that can take business context and goal into account, and generate campaigns. It also reacts to changes in goals or business context.
- Built a functional onboarding flow that enables the storage of context.
What we learned
- We'll have to build a couple of custom MCPs for precision.
What's next for Appents MAIA
- Make it working end-to-end!
- Integrate with platforms like Facebook, Twitter, IG, etc, for full ad buying automation
- Add support for more channels - email, sms campaigns, online marketplaces (amazon, shopify, etc) 4.Implement adaptive learning from campaign performance and user feedback
- Expand capabilities to enterprise workflows
- Enable businesses to bring in their data (through integrations, and files)
Built With
- google-cloud
- groq
- next
- node.js
- react
- toolhouse
- typescript
Log in or sign up for Devpost to join the conversation.