> Markdown version of [/jobs/ext/4294203-ai-ml-gcp-engineer](https://www.wearedevelopers.com/jobs/ext/4294203-ai-ml-gcp-engineer). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# AI/ML GCP Engineer

- **Company:** Saransh Inc
- **Location:** New York, NY, United States
- **Experience:** Experienced
- **Contract:** Temporary to permanent
- **Skills:** LangGraph Framework, Application Programming Interfaces (APIs), Artificial Intelligence, Computing Platforms, Data Governance, Memory Management, UN Electronic Data Interchange for Administration Commerce and Transport, Python (Programming Language), Machine Learning, Search Technologies, Model Garden (Google Cloud), Google Cloud, Fast Healthcare Interoperability Resources, Retrieval-Augmented Generation, Large Language Models, Multi-Agent Systems, Llamaindex, Agentic-AI, Fastapi, Google Cloud Functions, CrewAI, Health Level Seven International, AutoGen, Virtual Agents, Semantic Kernel, Serverless Computing, Docker, Microservices
- **Published:** October 30, 2026
- **Apply:** https://www.careerjet.com/jobad/usa39e8dee0ae82ad6e504e6e392af0d16

## About the Role

Core AI/ML & Agentic Engineering (Must-Have): 8+ years in software/AI engineering, with 3+ years directly building, deploying, and maintaining production-grade LLM applications, RAG pipelines, or autonomous agent frameworks.
Agentic Frameworks: Hands-on mastery of multi-agent orchestration patterns, tool calling, stateful graphs, and memory management (e.g., LangGraph, AutoGen, LlamaIndex, Semantic Kernel).
GCP AI Ecosystem: Deep experience with Vertex AI (Model Garden, Endpoint Deployment, Vector Search, Workbench) and cloud-native services (Cloud Run, Pub/Sub, Cloud Functions).
Production Python Engineering: Advanced Python expertise (AsyncIO, FastAPI, Pydantic, gRPC) writing clean, tested, and containerized microservices.
Healthcare / Payer Domain: Proven familiarity with Payer workflows (Prior Authorization, Claims Processing, Appeals, Member Engagement) and health data standards (FHIR, EDI X12, ICD-10/CPT).

## Description

Healthcare Payer Agentic Platform Architecture: Design and implement end-to-end multi-agent orchestration systems (using LangGraph, AutoGen, CrewAI, or custom DAG runners).
Forward Deployed Engineering & Client Value Delivery: Embed directly with engineering, product, and client business stakeholders to translate messy healthcare operational processes (e.g., prior auth rules, claims denial handling) into deterministic agent tools and graphs.
Production AI/ML Engineering on GCP: Build, deploy, and maintain high-throughput, low-latency microservices using Python, Docker, Cloud Run, and GKE.
Healthcare Data, Compliance & Security: Engineer AI agent tools that interface seamlessly with enterprise Payer data standards, including HIPAA-compliant environments, FHIR/HL7 APIs, and EDI transactions (837, 835, 270/271, 278).
Observability, Guardrails & Responsible AI: Deploy comprehensive telemetry to track agent trajectory, tool calling success rate, token usage, and sub-second reasoning latency using Google Cloud Operations.