Opportunity brief
Responsibilities:
-
Build AI agents and LLM-based applications in Python that automate real business workflows for our clients. -
Write durable, long-running workflows with Temporal.io, including retries, timeouts and human approval steps. -
Design prompts and structured outputs, and test them against real client data to make agent results reliable. -
Connect agents to client systems through APIs, databases and documents (emails, PDFs, spreadsheets). -
Build and maintain data pipelines that pull data from e-commerce platforms, marketplaces and APIs into structured datasets (secondary). -
Join client calls with senior engineers to understand requirements and explain how a solution works.
Skills & Experience:
Technical Skills:
-
Strong Python fundamentals: functions, classes, async code, error handling and testing. -
Hands-on work with at least one LLM API (Claude, OpenAI or Gemini), through a job, internship or personal project. -
A basic grasp of how AI agents work: tool calling, structured outputs and keeping state across steps. -
Interest in learning Temporal.io and agent frameworks such as PydanticAI or LangGraph. -
Basic understanding of ETL/ELT processes and REST APIs.
Soft Skills:
-
Clear written and spoken English, with the ability to explain technical ideas to non-technical people. -
Curious and quick to learn, with the habit of reading docs and trying things before asking. -
Takes ownership of a task and works through problems on their own.
Good to Have:
-
Postgres and solid SQL. -
Git, including branches and pull requests. -
Docker for running and shipping services. -
Experience using Claude Code, or writing Claude Code skills. -
Exposure to cloud platforms (AWS or GCP) or data warehouses (BigQuery, ClickHouse). -
A GitHub profile or demo of an agent or LLM project you built.