Convergehealth-full-stack Software Engineer-innovation_delivery_transformation

Deloitte Deloitte · Consulting

Full-Stack Software Engineer focused on designing, building, and scaling production-grade GenAI applications for the healthcare industry. The role involves working with modern front-end and backend frameworks, AI orchestration layers, RAG, agentic workflows, and multi-agent architectures within cloud-native environments.

What you'd actually do

  1. Design and implement full-stack applications using Angular or React on the front end and Node.js and Python-based services on the backend.
  2. Design and build agentic AI systems, including multi-agent coordination, tool invocation workflows, memory/state management, and structured reasoning pipelines.
  3. Implement AI-powered components such as RAG pipelines, embedding workflows, vector store integrations, and orchestration layers that connect LLMs to enterprise data and services.
  4. Deploy AI-enabled applications in AWS environments, leveraging managed services such as ECS/EKS, Lambda, S3, and Bedrock, while applying cloud-native infrastructure patterns to support scalable, production-grade systems.
  5. Integrate and operate open-source technologies across the stack, including model serving frameworks, vector databases, orchestration libraries, and containerized infrastructure components.

Skills

Required

  • Strong working knowledge of both JavaScript/TypeScript and Python in production environments.
  • Practical experience building and deploying cloud-native applications in AWS, with familiarity across compute, storage, API, and managed AI services.
  • Hands-on experience designing or experimenting with agentic systems using frameworks such as LangChain, LangGraph, CrewAI, Strands, LlamaIndex, DSPy, or similar.
  • Working understanding of multi-agent coordination patterns, tool integration, memory persistence, and structured reasoning workflows.
  • Familiarity with vector databases and embedding-based retrieval strategies in RAG architectures.
  • Comfort working across open-source ecosystems beyond models, including orchestration libraries, model serving stacks, vector stores, observability tooling, and modern infrastructure frameworks.
  • Strong debugging, experimentation, and system-level thinking in distributed cloud-native environments.

Nice to have

  • Forward-thinking engineering mindset, including embracing AI-enabled development tools (e.g., Claude Code, GitHub Copilot) as productivity multipliers.
  • Experience working effectively with globally distributed engineering teams across international time zones.
  • Familiarity with healthcare or life sciences industry concepts and data domains (e.g., claims, FHIR, EDI, remittance data, clinical datasets)
  • Intellectual curiosity and bias toward building, testing, and iterating quickly in ambiguous environments.

What the JD emphasized

  • agentic AI systems
  • multi-agent coordination
  • tool invocation workflows
  • RAG pipelines
  • vector store integrations
  • orchestration layers
  • cloud-native environments
  • agentic systems
  • multi-agent coordination patterns
  • tool integration
  • memory persistence
  • structured reasoning workflows
  • vector databases
  • embedding-based retrieval strategies
  • RAG architectures

Other signals

  • design and build agentic AI systems
  • implement AI-powered components such as RAG pipelines
  • deploy AI-enabled applications in AWS environments