Staff AI/Data Engineer

Superside

Completely RemoteFull TimeEngineering & Architecture
Posted Today

Job description

Responsibilities

  • Build the knowledge layer for Superads' AI agents to enable deep brand understanding and reasoning.
  • Design how brand knowledge is represented, stored, and continuously updated to mimic senior strategist judgment.
  • Develop agentic search capabilities that allow agents to identify knowledge gaps and navigate information dynamically.
  • Evaluate and iterate on architectures for knowledge retrieval and reasoning to improve reliability.
  • Translate emerging AI research into production-ready systems that enhance output quality.
  • Partner with product and AI engineering teams to ensure infrastructure improvements drive direct user impact.

Requirements

  • 8+ years of experience in high-performance engineering environments.
  • Proven track record of building agentic workflows and RAG systems in production.
  • Deep understanding of LLM knowledge representation, memory architectures, and reasoning under uncertainty.
  • Proficiency with Python and FastAPI.
  • Experience with LangChain or LangGraph for building multi-agent systems.
  • Hands-on experience with cloud data platforms such as Snowflake, BigQuery, or Databricks.

About the Company

Superside is a remote-first creative subscription platform. Through Superads, we are building AI agents that help marketers make better creative decisions by reasoning through strategy on their behalf. We are a global community spanning 60+ countries, dedicated to accelerating the world's transition to online work.

Skills & tools

PythonLangChainRAG

What the team is looking for

Use this list as a quick fit check before you apply.

  1. 018+ years high-performance engineering
  2. 02Experience building agentic workflows and RAG systems
  3. 03Deep understanding of LLM knowledge representation
  4. 04LangChain or LangGraph experience
  5. 05Strong Python and FastAPI skills
  6. 06Cloud data platform experience (Snowflake, BigQuery, or Databricks)