Founding Data & AI Engineer (Python / Django / LLMs), Creative Engine
Sierra Studio · Remote · Brazil (Remote) ·
COMPENSATION
€•• – ••k
Listed on the employer's job board (Ashby)
- Founding Engineer role
- Python, Django, LLMs
- Fully remote (Brazil)
- US time zones
- AI-assisted workflow
Join a performance creative agency as the founding engineer for their new Creative Engine. This data-first product centralizes insights from TikTok, Meta, and TikTok Shop to replace gut instinct with analytics. You will build the foundation that allows strategists to query video performance and trends directly.
The team operates asynchronously with high ownership, deploying to production multiple times daily. You will leverage AI-assisted coding tools to ship full-stack features, from data models to user interfaces, without dedicated frontend support.
What you'll do
- Design pipelines to structure performance and sales data across hundreds of accounts
- Build LLM analysis layers using Gemini for video tagging and Claude for generation
- Implement semantic search with pgvector to enable natural language queries
- Connect organic references, ad metrics, and creator history into a queryable system
- Develop creator-performance matching logic to improve casting decisions
- Build full-stack features including dashboards and export tools using Django and HTMX
- Integrate external APIs such as TikTok, Meta, and internal async pipelines
What you'll need
- 5+ years building production data-heavy applications with Python and Django
- Strong SQL skills including aggregations, joins, and window functions
- Experience designing and debugging async data pipelines with Celery and Redis
- Hands-on experience with LLM APIs, prompt engineering, and cost management
- Proficiency with agentic coding tools like Claude Code for rapid development
- Ability to integrate complex third-party APIs with rate limits and auth flows
- Comfortable building frontend interfaces with Django templates, HTMX, and Tailwind
- Strong written communication skills for an async, low-process team environment
Nice to have
- Experience building agent harnesses, RAG systems, or LLM evaluation pipelines
- Understanding of ML fundamentals and when to use simple scoring functions
- Background in video or audio processing pipelines
- Domain knowledge in ad tech, creator economy, or performance marketing
- Previous experience working with US-based teams in async settings
Stack
- Django
- PostgreSQL
- Celery
- Redis
- HTMX
- Tailwind CSS
- Google Gemini
- Anthropic Claude
- pgvector
- Docker
- AWS ECS
How you'll work
- Fully remote position
- Must be based in Brazil
- Work US time zones
- Asynchronous communication via Slack and GitHub
- No standups or status meetings