Junior Data Engineer - Data Tools

Allegro is a company developing in-house conversational analytics solutions including Looker Conversational Analytics (LCA) and Allegro Data Intelligence (ADI). Their solutions are used by non-technical business users and involve building production-ready conversational analytics tools leveraging Python, SQL, and Google Cloud Platform.

• Good foundations in Python and SQL from studies, internships, personal projects, or commercial experience.
• Basic understanding of data pipelines, APIs, data modeling, and software development practices.
• Basic understanding of or strong interest in large language models (LLMs) and agentic AI systems.
• Interest in learning how to test and monitor systems with non-deterministic outputs.
• Structured approach to debugging problems across multiple services and data sources.
• Willingness to take increasing ownership as knowledge of product and architecture grows.
• Good communication skills and a collaborative mindset.
• Communicative English at a minimum B2 level.

Nice to have:

• Experience with Google Cloud Platform.
• Familiarity with PostgreSQL, MongoDB, or other NoSQL databases.
• Experience with LLM application frameworks like LangChain, LangGraph, Langfuse.
• Basic knowledge of REST APIs, event-driven systems, or MCP.
• Familiarity with automated testing, CI/CD processes, GitHub Actions, or configuration-as-code.
• Basic understanding of authentication, authorization, and role-based access control.

• Develop and maintain Python services, APIs, and data integrations supporting LCA and ADI.
• Build pipelines that collect and process agent interactions, traces, feedback, and evaluation results.
• Work with SQL and NoSQL data models used by Conversational Analytics services.
• Help define, validate, test, and deploy conversational agents through repeatable, version-controlled workflows.
• Integrate agents with Looker, the corporate semantic layer, and other services through APIs or emerging standards such as MCP.
• Create automated regression tests and evaluations measuring factual correctness, consistency, and relevance.
• Improve observability and debugging of agentic workflows including model interactions, prompts, tool calls, and conversation traces.
• Document technical findings and contribute to code reviews and architectural discussions.