Middle/Senior AI Agent Developer — Беларусь, Минск

Description:

We are looking for a skilled AI Agent Developer to join our AI team. You should have proven production experience with AI systems and a strong interest in learning and growing with modern agent technologies.

Responsibilities

• Develop, implement, and optimize AI agents capable of autonomous decision‑making, reasoning, and tool‑usage in real‑world scenarios.

• Build and maintain pipelines for agent planning, memory, retrieval, and multi‑step task execution.

• Integrate LLMs, vector stores, APIs, and external tools to enable complex agent behaviors.

• Design and refine prompting strategies, reasoning chains, and agent policies to improve reliability and accuracy.

• Implement evaluation frameworks for agent performance, safety, and robustness across diverse tasks.

• Monitor agent behavior in production, troubleshoot failures, and continuously improve quality of outputs.

• Optimize inference pipelines for speed, cost efficiency, and model/tool selection.

Requirements

• Strong Python skills and proven production experience with Agentic ecosystems

• Embeddings & Vector search

• Prompt & Context engineering

• RAG, CRAG, GraphRAG

• LangChain, LangGraph, CrewAI

• Agent's testing & verification

• Agentic Tools and skills

• AI-assisted coding models & tools

• Strong analytical and problem-solving mindset

• Ability to clearly communicate findings and trade-offs

• Ownership of tasks from research to implementation

• Curiosity and willingness to explore new approaches

• English proficiency at A2 level or higher

Nice to have

• Autonomous Agents & Deep Agents

• Model types and capabilities (Thinking, Multimodal models)

• LLMOps

• GenAI Observability

• Memory management & Prompt caching

• Model Governance & Data Governance (Responsible AI)

• Basics knowledge of Cloud infrastructure

• Agentic patterns

• Integration mechanics (MCP, AA-protocol)

• Basic understanding of machine learning fundamentals:

• Supervised learning, models training, model evaluation

• Overfitting, regularization, cross-validation

• Knowledge of statistical methods and probability theory

We can offer:

• Projects for such clients as PayPal, Wargaming, Xerox, Philips, adidas and Toyota

• Competitive compensation that depends on your qualification and skills

• Career development system with clear skill qualifications

• Options to work remotely

• Compensation for medical expenses

• English courses online

• Corporate parties and events for employees and their children

• Internal conferences, workshops and meetups for learning and experience sharing

• Gym membership compensation, corporate sport competitions (cybersport included)

• 5 days of paid sick leave per year with no obligation to submit a sick-leave certificate