AI Solution architect

VySystems; New York (NY); Vor Ort; USD 180.000 - 240.000; Vollzeit

Zusammenfassung

VySystems is seeking an experienced AI Solution Architect for a full-time, on-site role in Manhattan, NY. The ideal candidate has 12–15 years designing enterprise AI solutions with expertise across major cloud platforms and secure, scalable architectures.

You will architect end-to-end ML, Generative AI, and RAG solutions, define agent architectures, select models and deployment patterns, and lead governance, security, and productionization efforts while mentoring teams.

Qualifikationen

• LLMs/SLMs, embeddings, vector search, Graph-RAG knowledge
• Python and AI frameworks: PyTorch, TensorFlow, Hugging Face, LangChain, LangGraph
• Cloud AI platforms and infra: Azure AI, AWS SageMaker, Google Vertex AI, Docker, Kubernetes, Terraform, CI/CD
• Databases and data pipelines: SQL/NoSQL, Kafka, Spark, Airflow, observability

Aufgaben

• Architect end-to-end ML, Generative AI, RAG, and Agentic AI solutions from data to governance.
• Define agent architectures covering orchestration, tools, context, memory, guardrails.
• Select models, frameworks, retrieval approaches, and deployment patterns by quality, latency, security, scale, cost.
• Design cloud-native and distributed systems using APIs, microservices, containers, and Kubernetes.
• Establish AI security and governance controls for privacy, access, regulatory needs.
• Lead discovery, assessments, proofs of concept, architecture reviews, and productionization; mentor teams.
• Apply Spec-Driven Engineering turning business intent into traceable specifications.

Kenntnisse

LLMs/SLMs

Embeddings

Vector search

Graph-RAG

Prompt engineering

Python

AI frameworks

LangChain/LangGraph

Tools

PyTorch

TensorFlow

Hugging Face

LangChain

LangGraph

Azure AI

AWS SageMaker

Google Vertex AI

Docker

Kubernetes

Terraform

CI/CD

Jobbeschreibung

We are seeking an experienced AI Solution Architect for a full-time, onsite role in Manhattan, NY. Candidates based in New York or New Jersey are preferred. The ideal candidate brings 12–15 years of experience designing and delivering secure, scalable, enterprise-grade AI solutions and expertise with at least one major cloud platform.

Responsibilities

• Architect end-to-end Machine Learning, Generative AI, RAG, and Agentic AI solutions, from data ingestion and model integration through deployment, evaluation, monitoring, and governance.
• Define agent architectures covering orchestration, tools and APIs, context and memory, human oversight, guardrails, and observability.
• Select models, frameworks, retrieval approaches, and deployment patterns based on quality, latency, security, scale, and cost.
• Design cloud-native and distributed systems using APIs, microservices, event-driven patterns, containers, and Kubernetes.
• Establish AI security and governance controls for privacy, access, responsible AI, model security, auditability, and regulatory needs.
• Lead discovery, assessments, proofs of concept, architecture reviews, reference architectures, and productionization; mentor teams and explain trade-offs to stakeholders.
• Apply Spec-Driven Engineering by turning business intent, architecture, requirements, interfaces, safeguards, and acceptance criteria into traceable, version-controlled specifications that guide implementation and testing.

Qualifications

Strong knowledge of LLMs/SLMs, embeddings, vector search, Graph-RAG, knowledge graphs, evaluation, and prompt/context engineering; hands-on Python and AI frameworks such as PyTorch, TensorFlow, Hugging Face, LangChain, or LangGraph; and experience with Azure AI, AWS SageMaker, or Google Vertex AI. Familiarity with SQL/NoSQL, vector and graph databases, Kafka, Spark, Airflow, MLOps/LLMOps, Docker, Kubernetes, Terraform, CI/CD, and observability is expected. Excellent client-facing communication, architecture documentation, stakeholder leadership, and experience with regulated industries are essential.