Generative Media Engineer, AI Social - US (Remote)

Luxury Presence is building the AI growth platform for real estate. Backed by Bessemer Venture Partners and other top investors, we're a Series C company that has hit $100M in annual recurring revenue. More than 90,000 real estate professionals, including over 30% of the WSJ Real Trends top 100 agents in the United States, use us to run and grow their business. The Opportunity AI Social is one of the most ambitious products we've ever shipped. An AI agent that produces on-brand Instagram carousels and reels for every Luxury Presence customer, from raw input data to finished media. It isn't LLMs writing code. It's LLMs designing, choosing the photo, voicing the script, picking the cuts, and composing the final frame. As the Generative Media Engineer on AI Social, you own the render pipeline, the prompts and agent loops that drive it, and the visual quality of what ships. Product and content strategy partners set the direction; engineering and design peers ship alongside you. This is an unusual role at the intersection of code, design, AI, and social media. You don't fit the normal engineering mold, that's the point. Skills and experience Solid working knowledge of TypeScript and Node.js. You're comfortable shipping production code and you're ready to grow into our stack. Hands-on experience with LLM-driven systems. Prompts, structured outputs, tool-calling, agent loops, evals. At least one shipped piece of generative media in image, video, voice, or music that someone other than you has seen and used. A portfolio or body of work that demonstrates both technical depth and design sensibility. Bonus: a personal creator practice somewhere. A feed, a portfolio, a side project, a YouTube channel, an Are.na. Public, ongoing, under your own name. You understand the medium because you've lived in it. Formal or self-taught grounding in design, visual art, film, or motion. Comfort moving between exploratory PoC work and shipped-to-customers polish, with judgment about which mode is appropriate when. Technical fluency with AI tools You're already using Claude Code, Cursor, or similar AI-assisted development tools in your daily workflow. We're a Claude Code shop and we expect you to be fluent. Familiarity with MCPs and how tools like Figma, Notion, and Linear connect into AI-assisted workflows. You understand how these tools and integrations fit together and can help the team get more out of them. Beyond fluency, you're a dabbler and experimenter. You try new prompts, push their limits, and share what you learn. You have opinions about where the medium is headed. Mindset You're a tinkerer. You don't just build what's asked; you spot leverage points and build what's needed. You care about systems more than features. You'd rather build a template family that ships ten posts well than ten bespoke posts that age out by next week. You build on shared systems. Constraints feel like leverage, not obstacles. You have strong convictions about quality, but you ship. You know the difference between polish that matters and perfection that stalls. You partner well with non-engineers. You'll collaborate across product, design, and content strategy. Patience and clarity with people outside engineering is non-negotiable. You operate with high ownership and low ego. You take direction from product and strategy partners on what to build, and you raise the bar on everything you ship. What You Won't Be Doing To be clear about what this role is not: You won't own platform architecture or distributed systems work. You'll partner with engineering teams that do, while you stay focused on the AI Social pipeline itself. You won't own the social strategy. You'll have a real voice in shaping it, but the call sits with our product and content strategy partners. You won't be managing people on day one. This is a hands-on IC role with the expectation that you lead through craft and influence. Tech stack Platform: Node, TypeScript, React, Postgres, AWS, Temporal LLMs and agents: the major model SDKs, with tool-calling, structured outputs, and multi-step agent loops Media pipeline: image and video generation (Veo for video), voice and music synthesis (ElevenLabs), composite (Shotstack), and headless browser rendering