Senior AI-Native Full-Stack Software Engineer

EPAM Systems; Poland; On-site; PLN 250,000 - 360,000; Full time

Job summary

EPAM Systems seeks a Senior AI‑Native Full‑Stack Software Engineer to industrialize AI prototypes into production-grade services. You will build agentic systems spanning Python, Java Spring Boot, and TypeScript/Angular, with strong emphasis on eval, security, and observability across the stack.

Office in Warsaw required 3 days per week. You will work on Azure-based platforms, RAG pipelines, vector databases, and knowledge graphs, delivering end-to-end solutions with robust CI/CD and cost-aware

Qualifications

• 5+ years of experience shipping LLM/agentic systems in production with real users, backed by defined eval approaches and demonstrated scale.
• Proficiency in Python, Java Spring Boot, and TypeScript/Angular for full-stack delivery.
• Expertise in Azure PaaS, Azure AI Foundry, and Azure OpenAI services.
• Knowledge of agentic frameworks, vector databases, and knowledge graphs, combined with MCP-based tool integration.
• Skills in prompt and context engineering.
• Background in CI/CD pipelines and observability practices, having operated production systems end-to-end.
• Familiarity with AI-assisted engineering workflows, coding daily with AI agents and demonstrating the workflow live.
• Understanding of evaluation-first development, ensuring no AI capability ships without a measured baseline and regression eval.
• Competency in cost and latency optimization, treating token economics as a core engineering feature.
• Understanding of regulated-environment discipline, including data classification, auditability, and least privilege principles.
• English proficiency at B2 level or higher

Responsibilities

• Industrialize prototypes into production services on Azure AI Foundry and Azure OpenAI, moving from build-ready packs to systems real users depend on within weeks.
• Build agentic systems using orchestration frameworks, RAG pipelines, vector databases, knowledge graphs, and MCP-based tool integration, chosen by need and engineered for change.
• Establish the eval harness first, with golden sets, regression evals, and guardrail tests wired into CI to measure quality on every change.
• Engineer guardrails including input/output filtering, grounding and citation, PII protection, rate limits, and human escalation paths.
• Deliver full-stack solutions across Python, Java Spring Boot services, and TypeScript/Angular front ends without relying on separate ownership layers.
• Run production engineering end-to-end, covering CI/CD, observability with traces on every LLM call, and cost and latency management.
• Absorb model-version churn by design, ensuring systems remain resilient as underlying models evolve.
• Build security and compliance into the system, respecting data classification boundaries in prompts, stores, and logs, externalizing secrets, and ensuring every AI decision is auditable.
• Iterate from real usage through hypercare, tuning, and fixes based on evidence, then package successful patterns for future use.

Skills

Full-stack development

Python

Java Spring Boot

TypeScript/Angular

Azure services

CI/CD

Observability

Security and compliance

Evaluation-first development

English communication

Tools

Azure PaaS

Azure AI Foundry

Azure OpenAI

Vector databases

Knowledge graphs

RAG pipelines

Job description

We are seeking a Senior AI-Native Full-Stack Software Engineer to industrialize AI prototypes into production-grade services, building agentic systems that real users depend on while embedding evaluation, security, and observability into every layer of the stack. Please note that working from the office in Warsaw 3 days per week is required.

Responsibilities

• Industrialize prototypes into production services on Azure AI Foundry and Azure OpenAI, moving from build-ready packs to systems real users depend on within weeks
• Build agentic systems using orchestration frameworks, RAG pipelines, vector databases, knowledge graphs, and MCP-based tool integration, chosen by need and engineered for change
• Establish the eval harness first, with golden sets, regression evals, and guardrail tests wired into CI to measure quality on every change
• Engineer guardrails including input/output filtering, grounding and citation, PII protection, rate limits, and human escalation paths
• Deliver full-stack solutions across Python, Java Spring Boot services, and TypeScript/Angular front ends without relying on separate ownership layers
• Run production engineering end-to-end, covering CI/CD, observability with traces on every LLM call, and cost and latency management
• Absorb model-version churn by design, ensuring systems remain resilient as underlying models evolve
• Build security and compliance into the system, respecting data classification boundaries in prompts, stores, and logs, externalizing secrets, and ensuring every AI decision is auditable
• Iterate from real usage through hypercare, tuning, and fixes based on evidence, then package successful patterns for future use

Requirements

• 5+ years of experience shipping LLM/agentic systems in production with real users, backed by defined eval approaches and demonstrated scale
• Proficiency in Python, Java Spring Boot, and TypeScript/Angular for full-stack delivery
• Expertise in Azure PaaS, Azure AI Foundry, and Azure OpenAI services
• Knowledge of agentic frameworks, vector databases, and knowledge graphs, combined with MCP-based tool integration
• Skills in prompt and context engineering
• Background in CI/CD pipelines and observability practices, having operated production systems end-to-end
• Familiarity with AI-assisted engineering workflows, coding daily with AI agents and demonstrating the workflow live
• Understanding of evaluation-first development, ensuring no AI capability ships without a measured baseline and regression eval
• Competency in cost and latency optimization, treating token economics as a core engineering feature
• Understanding of regulated-environment discipline, including data classification, auditability, and least privilege principles
• English proficiency at B2 level or higher