Generative AI (GenAI) Architect

Role : Generative AI (GenAI) Architect

Location : Charlotte, NC or Dallas, TX (Hybrid)

Role Overview

We are seeking an experienced Generative AI Architect to design, build, and operationalize enterprise-grade GenAI solutions with a strong focus on Python development and Identity & Access Management (IAM) automation. You will lead architecture across LLM platforms, RAG pipelines, LLMOps, and security/governance, while automating IAM controls and policies across Azure and/or AWS to ensure secure, scalable, and cost-efficient deployments.

Key Responsibilities

• Architecture & Design

• Define endtoend GenAI architectures: model/platform selection, prompt orchestration, RAG, agents, observability, guardrails, and cost governance.

• Produce architecture artifacts (HLD/LLD), security patterns, data access boundaries, and reference implementations.

• Hands-on Engineering (Python)

• Build APIs/services (FastAPI/Flask), orchestration layers, SDK integrations (Azure OpenAI, AWS Bedrock, OpenAI), and reusable libraries for prompt templates, retrievers, evaluators, and logging/tracing.

• Implement data ingestion, chunking strategies, embeddings, and vector databases (FAISS, Pinecone, Milvus, Weaviate).

• LLMOps / MLOps

• Establish CI/CD for GenAI workloads; prompt/version management; automated evaluation, telemetry, A/B testing, red-teaming, and rollback strategies.

• Integrate observability (tracing, logging, metrics) and reliability mechanisms (rate limiting, circuit breakers, caching).

• Security & IAM Automation

• Design and implement secure identity and access management strategies for GenAI workloads.

• Automate IAM processes for user, group, and role provisioning across cloud environments using Infrastructure-as-Code and scripting.

• Establish policy-driven access controls (RBAC/ABAC), single sign-on (SSO), and identity federation for seamless integration.

• Implement secrets management, key rotation, and audit logging to ensure compliance and governance.

• Drive automation for lifecycle management, conditional access, and enforcement of least-privilege principles.

• Performance & Cost Optimization

• Benchmark models; optimize context windows, prompt strategies, retrieval quality; tune throughput/latency and total cost of ownership (autoscaling, caching, batch inference).

• Stakeholder Leadership

• Lead client workshops, translate business outcomes into technical roadmaps, manage risk, and deliver executive readouts/demos.

Required Skills & Experience

• 8 12 years in software/data/platform engineering; 3 5 years in AI/ML with 1 3 years in Generative AI architecture and delivery.

• Strong Python: FastAPI/Flask, pandas, asyncio, packaging/testing, SDKs for LLMs/vector DBs; experience building production-grade APIs and services.

• LLM Platforms: Azure OpenAI, AWS Bedrock, OpenAI API, Vertex AI, or Databricks (at least one at enterprise scale).

• RAG Expertise: embeddings, chunking, retriever optimization; hands-on with FAISS/Pinecone/MilvWeaviate.

• Cloud & DevOps: Azure and/or AWS (IAM, networking, storage, serverless, containers), Docker/Kubernetes, CI/CD (GitHub Actions/Azure DevOps), IaC (Terraform/ARM/CloudFormation).

• IAM Automation (Must-have)

• Strong experience in designing and automating identity and access management processes across enterprise environments.

• Ability to implement role-based and policy-driven access controls, single sign-on (SSO), and identity federation.

• Hands-on experience with automating user, group, and role provisioning using Infrastructure-as-Code and scripting.

• Expertise in secrets management, key rotation, and enforcing least-privilege principles.

• Familiarity with compliance-driven IAM practices, audit logging, and lifecycle management automation.

• Security & Compliance: AI safety, guardrails, content moderation, hallucination mitigation, and alignment to SOC 2/HIPAA/PCI-DSS (client-dependent).

• Communication: Excellent client-facing communication, documentation, and stakeholder management.

Preferred Qualifications

• Enterprise search (Elastic, Azure Cognitive Search, OpenSearch) and document intelligence/OCR (Azure Document Intelligence, Amazon Textract).

• Model tuning (prompt tuning, LoRA, selective fine-tuning) and evaluation frameworks (automated human-in-the-loop).

• Observability tooling: OpenTelemetry tracing for LLM pipelines, Datadog/New Relic.
• Experience in regulated industries (Financial Services, Healthcare, Telecom)