DataJobs.io
← Back to all jobs

Job Description

Build production-grade, AI-first agentic systems with LLMs in a role that focuses on multi-agent orchestration, retrieval augmented generation (RAG), tool calling, and enterprise deployment. This onsite position in Austin, TX supports the full lifecycle, from system design and implementation to testing, observability, and on-call support to keep services available, reliable, and efficient.

What you’ll do

  • Partner with stakeholders to interpret requirements, validate assumptions, and incorporate feedback into evolving designs and solution fixes.
  • Contribute to system design and architecture by refining code plans and maintaining design documentation.
  • Develop, implement, and maintain product code, services, and components using established coding patterns, guidelines, and best practices.
  • Use debugging tools to verify assumptions and proactively flag issues before they impact production.
  • Participate in code reviews to uphold coding standards and handle routine pull requests.
  • Create test plans and configure testing procedures to identify and resolve defects across multiple features.
  • Support on-call rotations by troubleshooting issues and deploying fixes under guidance when needed.
  • Build and tune multi-agent LLM workflows using orchestration frameworks such as LangGraph, LangGraph4j, and LangChain4j, including declarative graph compilation, conditional routing, state persistence, and dynamic spawning at runtime.
  • Implement ReAct agent loops where an LLM iteratively selects tools, interprets tool results, and continues until a terminal answer is produced.
  • Develop Supervisor/Orchestrator agent patterns where a Lead Agent coordinates Specialist Sub-Agents using conditional graph edges and tool frameworks.
  • Build MCP (Model Context Protocol) tool servers that expose enterprise data sources as callable tools for AI agents.
  • Implement and improve RAG pipelines covering document ingestion (PDF/Word), chunking, embedding generation, vector storage (for example PostgreSQL with pgvector), and semantic retrieval to ground LLM responses.
  • Integrate with enterprise LLM inference gateways via REST APIs, including request shaping, prompt template management, token budget control, and graceful degradation.
  • Design agent memory architectures with short-term per-agent context stores and long-term shared knowledge repositories.
  • Add confidence scoring and uncertainty quantification, including threshold-based routing for low-confidence results to human review.
  • Implement agent feedback loops that capture human reviewer decisions and feed structured corrections back into prompt templates and knowledge repositories.
  • Write and wire LangChain4j/LangChain tool definitions (such as Java @Tool-annotated methods) into ReAct agents for data transformation and external API calls.
  • Create agentic audit trails that log tool calls, agent decisions, prompts and responses, and state transitions for traceability and explainability.
  • Develop and test Spring Boot microservices (Java 21, Spring Boot 3.5.x) that host orchestration engines, expose REST APIs, and integrate with upstream data systems.
  • Author Helm charts and Jenkins CI/CD pipelines for containerized deployment to Kubernetes/OpenShift across multi-region on-premise data centers.
  • Build end-to-end and integration tests for non-deterministic AI components, including prompt regression suites, tool call mock frameworks, schema validators, and determinism gates.

What you bring

  • 2 or more years of work experience with a Bachelor’s Degree or an Advanced Degree (e.g., Masters, MBA, JD, MD, PhD).
  • 3 or more years of work experience with a Bachelor’s Degree, or more than 2 years of work experience with an Advanced Degree (e.g., Masters, MBA, JD, MD).
  • 2+ years of relevant experience and a Bachelor's degree, or 5+ years of relevant work experience.
  • Experience developing and/or implementing web-based or mobile applications (minimum two years) and familiarity with system design and architecture for product components.
  • Experience debugging and troubleshooting software issues, participating in code review, and applying coding standards.
  • Experience with test planning and execution for software features, and responding to support requests by deploying fixes.
  • Backend experience in Java (Java 17+ preferred; Java 21 a strong plus), including REST API design, Spring Boot, and JPA/ORM-based data access patterns.
  • Foundational understanding of LLM APIs including prompt construction, token limits, and response parsing.
  • Experience building and testing enterprise-scale web services (minimum one year) and working on client-facing project or technical teams (minimum one year).
  • Experience integrating feedback into design and solution fixes, plus collaborating cross-functionally and mentoring junior engineers.
  • Hands-on experience building agentic AI systems using LangGraph, LangChain, LangGraph4j, or LangChain4j, including multi-agent graph construction, node and edge definitions, conditional routing, and state schema design.
  • Experience implementing the React prompting pattern (Thought Action
  • Observation) in a production or near-production LLM application.
  • Working knowledge of MCP (server registration, tool schema definition with tools/list and tools/call, and client-side integration).
  • Experience building RAG pipelines including chunking strategies, embedding models, vector database querying (pgvector, Pinecone, Weaviate, or equivalent), and retrieval relevance tuning.
  • Experience with prompt engineering such as structured output enforcement (JSON schema), chain-of-thought prompting, few-shot example design, and system prompt management.
  • Experience with LLM observability including token usage, latency per agent step, tool call success rates, and output quality metrics in production.
  • Experience with PostgreSQL including the pgvector extension for embedding storage and similarity search.
  • Experience deploying containerized workloads on Kubernetes or OpenShift, including Helm chart authoring, rolling deployments, and health probes.
  • Experience designing audit logging for AI agent decisions, and familiarity with Aspect-Oriented Programming (AOP) in Spring Boot for cross-cutting concerns such as logging, latency measurement, and security enforcement.
  • Experience with CI/CD pipelines (Jenkins or equivalent) for AI/ML service deployments, including version gating and environment promotion strategies.

Compensation

USD 110,700 - 171,800 per year.

Benefits

  • Medical
  • Dental
  • Vision
  • 401(k)
  • FSA/HSA
  • Life Insurance
  • Paid Time Off
  • Wellness Program

Work location and logistics

Austin, TX (onsite). Work hours vary based on departmental needs. Travel requirement: 5-10% of the time. The role is performed in an office setting and involves sitting and standing at a desk, communicating in person and by telephone, and frequently operating standard office equipment such as telephones and computers.

Similar Jobs