Senior Software Engineer - Backend/platform Agentic AI

Mastercard Mastercard · Fintech · Arlington, VA +1 · AI & Data

Senior Software Engineer focused on building and operating a first-party AI platform with agentic, conversational, and generative AI capabilities for enterprise analytics products. The role involves end-to-end development of agentic AI systems, defining technical direction, building AI-enabled services, and ensuring production-grade infrastructure, scalability, reliability, and adherence to governance and Responsible AI standards within a regulated environment.

What you'd actually do

  1. Lead end-to-end development of agentic AI systems from design through production. This includes orchestration, tool calling, context engineering, retrieval, and streaming responses.
  2. Define technical direction for AI capabilities within the PI platform, driving architecture, design patterns, and integration strategies
  3. Build and operate AI-enabled services in Java and Python within a multi-tenant, customer-facing environment, ensuring scalability, reliability, and strict data isolation
  4. Design and implement production-grade AI infrastructure, including prompt management, evaluation frameworks, guardrails, observability, and cost/token telemetry
  5. Partner with platform teams (agent frameworks, LLM gateway), vendors (semantic data layer), and product teams to deliver integrated, end-to-end solutions

Skills

Required

  • Expertise with Java for backend services (Spring Boot, microservices)
  • Fluent in Python for AI/ML development (agentic frameworks, scripting, integrations)
  • Hands-on experience building LLM-powered production systems (API integration, prompt management, streaming, error handling, cost management)
  • Experience with agentic frameworks (LangGraph, LangChain, or similar), RAG pipelines, or AI orchestration systems
  • Proven ability to design scalable distributed systems with strong observability (logging, metrics, tracing, alerting)
  • Experience with CI/CD and modern SDLC practices (automated testing, quality gates, deployment automation)
  • Cloud experience (AWS or Azure), including managed AI/ML services
  • Technical leadership in design reviews, mentoring, and setting engineering standards
  • Solid backend/software engineering experience with ownership of distributed systems in production

Nice to have

  • Experience with multi-tenant architectures and customer data isolation
  • Familiarity with AI evaluation frameworks (agent evaluation, prompt regression testing, output quality metrics)
  • Experience with Databricks, Snowflake, or similar data platforms
  • Knowledge of vector databases, semantic search, knowledge graphs, and MCP
  • Exposure to analytics platforms, BI tools, or semantic data layers
  • Understanding of AI governance and Responsible AI practices in regulated environments

What the JD emphasized

  • production AI systems
  • agentic AI systems
  • tool calling
  • RAG
  • orchestration
  • production-grade AI infrastructure
  • guardrails
  • observability
  • Responsible AI standards
  • regulated environment
  • evaluation frameworks

Other signals

  • production AI platform
  • agentic, conversational, and generative AI capabilities
  • natural-language analytics
  • automated report summaries
  • personalized dashboard experiences
  • production AI systems
  • AI solutions from architecture through production at enterprise scale
  • agentic AI systems
  • orchestration, tool calling, context engineering, retrieval, and streaming responses
  • AI-enabled services
  • production-grade AI infrastructure
  • prompt management, evaluation frameworks, guardrails, observability, and cost/token telemetry
  • LLM gateway
  • engineering standards for AI development
  • Responsible AI standards
  • continuous improvement
  • task success rate, latency, cost per interaction, human intervention rate
  • evaluation coverage
  • productionizing AI/ML systems
  • building agentic or LLM-based systems
  • tool/function calling, RAG, context management, prompt engineering, and orchestration
  • operational ownership mindset
  • observability, incident response, and service reliability
  • pragmatic architectural decisions
  • LLM-powered production systems
  • agentic frameworks
  • RAG pipelines
  • AI orchestration systems
  • scalable distributed systems with strong observability
  • managed AI/ML services