6 days ago
Base Salary
$111k - $172k/yr
Responsibilities
- Collaborate with stakeholders to interpret requirements, refine designs, estimate work, incorporate feedback, and escalate risks.
- Design, implement, debug, review, test, maintain, and support software components and services.
- Design and tune multi-agent LLM workflows using LangGraph, LangGraph4j, LangChain, and LangChain4j.
- Build ReAct agent loops, supervisor/orchestrator patterns, MCP tool servers, agent memory, feedback loops, confidence scoring, and structured audit trails.
- Develop RAG pipelines covering document ingestion, chunking, embeddings, vector storage, and semantic retrieval.
- Develop Java and Spring Boot microservices that host AI orchestration engines and expose REST APIs.
- Create end-to-end and integration tests for prompts, tool calls, agent outputs, schemas, and determinism.
- Author Helm charts and maintain Jenkins pipelines for containerized deployments to Kubernetes/OpenShift across on-premise data centers.
- Participate in on-call rotations, troubleshoot production issues, deploy fixes, and improve reliability, observability, efficiency, and performance.
Requirements
- At least 2 years of work experience with a bachelor’s degree or advanced degree.
- Preferred experience includes 3+ years with a bachelor’s degree, more than 2 years with an advanced degree, or 5+ years of relevant work experience.
- At least 2 years of experience in a directly related technology or software field and in developing or implementing web-based or mobile applications.
- Experience with system design, architecture, debugging, code reviews, coding standards, test planning, support requests, and software developer tools.
- Experience developing backend services in Java, preferably Java 17 or later, with Java 21 strongly preferred.
- Familiarity with REST API design, Spring Boot, JPA/ORM data access, and enterprise-scale web services.
- Foundational knowledge of LLM APIs, prompt construction, token limits, and response parsing.
- At least 1 year of experience building and testing enterprise-scale web services and at least 1 year working on client-facing project or technical teams.
- Experience building agentic AI systems with LangGraph, LangChain, LangGraph4j, or LangChain4j, including multi-agent graphs, conditional routing, and state schemas.
- Experience implementing the ReAct prompting pattern in a production or near-production LLM application.
- Working knowledge of Model Context Protocol, including server registration, tool schemas, and client integration.
- Experience building RAG pipelines, using embedding models and vector databases such as pgvector, Pinecone, or Weaviate, and tuning retrieval relevance.
- Experience with prompt engineering, structured JSON output, chain-of-thought prompting, few-shot examples, and system prompt management.
- Experience monitoring LLM observability metrics such as token usage, agent-step latency, tool-call success rates, and output quality.
- Experience with PostgreSQL and the pgvector extension, Kubernetes or OpenShift deployments, Helm charts, audit logging, AOP, and Jenkins or equivalent CI/CD pipelines.
- Experience mentoring junior engineers and collaborating with cross-functional teams.
Benefits
- Medical, dental, vision, 401(k), FSA/HSA, life insurance, paid time off, and wellness program.
- At least 3 days per week in the office, with expectations confirmed by the Hiring Manager.
- Candidates must be able to participate in onsite interviews at Visa’s Austin, Texas office.
- The position requires 5–10% travel.
- Work hours vary according to department needs.
Tech Stack
Categories
About Visa
Visa builds and operates a global digital payments network that connects consumers, merchants, financial institutions, and governments. It earns fees for processing card transactions and offering risk, data, and value‑added services through VisaNet and products like tokenization and tap‑to‑pay. Founded in 1958 and headquartered in Foster City, California, Visa is a public company (NYSE: V) serving clients in more than 200 countries and territories.
