Oksana Novak

Recruiter

Oksana Novak

onovak@binariks.com

[email protected]

Agentic Workflow Engineer in Remote (Ukraine)

Binariks is looking for a highly motivated and skilled Agentic Workflow Engineer  to join a creative and strategic growth startup environment focused on transforming the commercial Property & Casualty (P&C) insurance space.


About the project: 

The core objective is to build and launch a cutting-edge AI application by orchestrating teams of AI agents. This application will surface new connections and insights across critical disciplines like underwriting, distribution, and compliance, and beyond.

What We’re Looking For


  • 4+ years of working experience in Data Science, ML, etc.


  • Experience with Agent Orchestration frameworks, specifically LangGraph and/or LangChain, and Pydantic AI, etc.


  • Proficiency in Model Context Protocol (MCP) server development.


  • Strong knowledge of Enterprise Python libraries (Pydantic, Django, SQLAlchemy, etc.). 


  • BS in Computer Science, Engineering, or a related discipline.


  • B2 + level of English


Will be a plus


  • Experience with SQL, Redis, Graph, and Vector database technologies.


  • Experience with Azure Microservice/Serverless architectures.


  • Knowledge of DevOps practices including Test Design, Version Control, CI/CD, Infrastructure as Code (IaC), etc.


  • Experience with Web UI frameworks (Svelte, React, Streamlit, etc.). 


  • MS in Computer Science, Engineering, or a related discipline.



Your Responsibilities:


  • Agent Construction: Construct AI agents using prompt and context engineering.


  • Logic Containment: Constrain agent behavior with conditional business logic.


  • Data Provisioning: Arm agents with static and dynamic data sources.


  • Workflow Orchestration: Orchestrate multi-agent teams to handle complex tasks.


  • System Design: Design and maintain modular architectures that connect agents to internal APIs, document stores, and analytics pipelines.


  • Testing & Evaluation: Automate system evaluation/testing methodology against quality benchmarks.


  • Launch an MVP multi-agent workflow that demonstrably improves internal process efficiency.


  • Establish an internal tooling and documentation to scale AI workflows across teams.


  • Collaborate with product and domain experts to continuously improve agent accuracy and reliability.