Přejít na hlavní obsah

Python Developer – AI Agents (Junior/Mid)

Technology
Sirius42
Česko, Česká republikapřed 5 dny

Popis pozice

Build AI agents on Google Cloud with a lead architect. Czech or Slovak speakers, fully remote within Czechia.

At Sirius42 we build and deploy AI agents for our clients on Google Cloud and Gemini Enterprise. We're adding a junior or mid-level Python developer, with the level set by your experience.

How we work: contract (IČO), remote from anywhere in Czechia, with visits to our Brno or Prague offices or travel to customer premises from time to time.

What you'll do

  • Build agents in Python with Google's Agent Development Kit (ADK): tools, prompts and multi-agent flows
  • Connect agents to company data: Vertex AI Search, Google Drive, BigQuery and internal APIs
  • Deploy agents to Vertex AI Agent Engine and register them in Gemini Enterprise
  • Write tests and evaluations so agents stay accurate when prompts or models change
  • Help set up Gemini Enterprise: data connectors, permissions and agent configuration

What you need

  • Working Python: functions, classes, error handling and basic async. You can read and extend someone else's code.
  • A basic grasp of LLM agents: prompts, tool calling, context limits and why agents hallucinate. You've built at least one thing on an LLM API, and side projects count.
  • Testing: unit tests (pytest or similar), plus an understanding of agent evaluations. That means a set of test questions with expected answers or tool calls, rerun after every change and scored rather than marked pass/fail.
  • Cloud basics, with Google Cloud preferred: service accounts, IAM roles, deploying to a managed service, keeping secrets out of code and reading cloud logs.
  • REST APIs and JSON: calling APIs, handling errors and getting started from SDK docs on your own.
  • Git, pull requests, code review, the command line, virtual environments and environment variables.
  • Czech or Slovak. You can read English technical docs; spoken English isn't required.

Nice to have

  • Vertex AI, Cloud Run, BigQuery
  • Google ADK, LangChain/LangGraph or another agent framework
  • ADK evaluation or the Vertex AI Gen AI evaluation service
  • Gemini Enterprise. It's new, so we don't expect experience.
  • RAG and vector search
  • MCP or A2A protocols
  • Docker, Terraform, uv, pydantic
  • Working directly with clients

What you get: real agent work on the newest Google AI stack, a lead architect to learn from, and room to grow from junior to mid and beyond.

Apply via LinkedIn with your CV and a link to something you've built with an LLM, such as GitHub or a side project.

Keywords
JSONPythonUnit TestingGoogle DriveBigQuery

Zajímá vás tato pozice?