Description
Summary Hey, We're building an AI-powered social media automation platform and looking for a team who can actually pull it off end-to-end. The idea is simple to describe but not simple to execute — one intelligent dashboard that handles the entire content lifecycle for creators, marketers, and agencies. Ideation, generation, scheduling, publishing, insights. All in one place, powered by AI that learns what works and gets sharper over time. There are a lot of tools that do pieces of this. What we want to build is the one product that does all of it properly, so users stop bouncing between five different tabs to run their content calendar. What we want the product to do At the core, it's a multi-modal AI engine — text, visuals, and short-form video — that generates content tailored to each platform. LinkedIn shouldn't sound like TikTok, and the product should just know that. Users refine tone and style with AI credits, then schedule and publish across channels without ever leaving the dashboard. On top of that, we need smart scheduling so calendars basically run themselves, workflow automation for approvals and brand voice consistency, and an analytics layer that feeds engagement data back into the AI so output quality keeps improving. For agencies and growing teams, we'll need workspaces, team roles, high-volume posting, and clean client management. Nothing here is wildly experimental. The hard part is making all of it feel like one connected system instead of five bolted-together tools — which is where most attempts at this fall apart. That's what we care most about getting right. A few things we'd love to hear your take on Rather than us dictating the stack, we'd like to see how you'd approach it. In particular: How would you architect the multi-modal AI layer so we're not locked into one model provider and can control costs as usage scales? What's your plan for social publishing reliability across Meta, LinkedIn, X, TikTok, and YouTube — including the API approval processes? How would you design the credit-based usage system so it holds up under both solo creators and agencies pushing hundreds of posts a month? How would you handle the feedback loop from analytics back into the generation engine? We're open on stack — Next.js frontend and a Python or Node backend feels like a natural fit, but if you've shipped something similar and have strong opinions, we want to hear them. Who we're hoping shows up Honestly, we're looking for a team that's done something adjacent to this before. Not "AI experience" in general — specifically people who've built AI content tools, social scheduling platforms, or marketing automation SaaS and know where the real sharp edges are. If you've been through the Meta or TikTok API approval processes, that alone tells us a lot. We'd rather work with a small team with strong product judgment than a big team taking orders. If you push back on something because you've seen it fail before, that's a green flag for us. How this would work Milestone-based on Upwork, weekly demos, async updates via Loom or similar. v1 is the core create-schedule-publish loop working end-to-end. v2 layers in the analytics feedback loop, video generation at scale, and the full agency feature set. If v1 lands well, we're very interested in a longer engagement — this isn't a one-and-done build for us. Timeline and budget we'd like to finalize on a short call once we understand each other properly. We're not chasing the cheapest quote and won't pretend otherwise. If you want to reply Please don't send a template. We can tell, and we'll skip it. Instead, tell us: One product your team has shipped that's genuinely close to this — AI content, social automation, marketing SaaS. Link it, tell us what your team owned, and one problem you had to actually think hard about. Your take on the biggest architectural risk on a build like this. One paragraph is enough. Which parts of the scope you'd own confidently on day one, and which parts you'd want to scope more carefully before committing to numbers. Who's on your team, roughly when you could start, and how you prefer to work. Looking forward to seeing what you've built. If your work is genuinely close to what we're going for, this should be a fun one to work on together. Thanks,