Medior Data/ai Engineer

AT&T AT&T · Telecom · CZE:642:Brno +2

Medior Data/AI Engineer role focused on developing and supporting intelligent systems using LLMs, prompt engineering, and data engineering best practices. The role involves building, integrating, and maintaining AI-powered applications and workflows, automating processes with LLMs, RAG, and AI agents, and working with frameworks like LangGraph. Responsibilities include prompt design, context management, integration testing, conversational/agent workflow development, and backend service maintenance.

What you'd actually do

  1. Assist with designing and optimizing prompts for LLMs across relevant use cases.
  2. Help implement context management strategies (e.g., session memory, retrieval-augmented generation).
  3. Support integration and testing of LLMs within Python-based applications using APIs and standard data pipelines.
  4. Participate in building, testing, and maintaining conversational and agent workflows (e.g., with LangGraph).
  5. Help maintain backend services, APIs, and data pipelines to ensure reliability and performance.

Skills

Required

  • Python development (FastAPI, Flask; SQL/NoSQL databases)
  • Prompt engineering for LLMs (e.g., OpenAI, Anthropic, open source)
  • Context management concepts (session management, vector search, retrieval)
  • Agentic/conversational flow frameworks such as LangGraph
  • Retrieval-Augmented Generation (RAG) and multi-agent workflow concepts

Nice to have

  • cloud environments
  • basic containerization (Docker)
  • modern API development and microservice patterns (REST, GraphQL)
  • data engineering tools/technologies (e.g., Spark, Kafka, Databricks, Snowflake) and SQL/another relevant language
  • AI-powered IDEs and tools (e.g., Visual Studio, GitHub Copilot, Windsurf, Cursor)
  • fine-tuning language models
  • statistics, machine learning (probability, hypothesis testing, experimentation, regularization)
  • predictive model building and validation (e.g., feature engineering, model training, error analysis)
  • LLM evaluation strategies (including automated/human-in-the-loop, task quality, safety metrics)
  • RAG-focused data science, prompt optimization, and LLM troubleshooting
  • prompt optimization, adaptation techniques (LoRA/QLoRA, DPO/IPO), data labeling, and evaluation approaches
  • basic system performance considerations for LLMs (token optimization, latency, model routing)
  • interpretability and debugging practices for GenAI/AI systems

What the JD emphasized

  • actively contribute to developing and supporting intelligent systems built around large language models (LLMs), prompt engineering, and data engineering best practices
  • help build, integrate, and maintain AI-powered applications and workflows
  • automating workflows using LLMs, Retrieval-Augmented Generation (RAG), and next-gen AI agents
  • building, testing, and maintaining conversational and agent workflows (e.g., with LangGraph)
  • Experience with agentic/conversational flow frameworks such as LangGraph
  • Experience with Retrieval-Augmented Generation (RAG) and multi-agent workflow concepts

Other signals

  • LLM
  • RAG
  • Agentic workflows
  • LangGraph
  • Prompt Engineering