AI Solution Architect

KData Inc.; United States; Remote; USD 180,000 - 250,000; Full time

Job summary

KData Inc. seeks an experienced AI Architecture Lead to own end-to-end AI delivery, from discovery to production monitoring and retirement.

You will partner across business, product, data science, engineering, operations, security and privacy to deliver measurable value through architectural decisions and hands-on validation. Responsibilities include translating business needs into AI use cases, designing data and knowledge pipelines, and evaluating build vs buy options.

Qualifications

• Typically 10+ years of relevant technology experience with architecture responsibility.
• End-to-end architecture spanning models, data, retrieval, apps, cloud, security and ops.
• Strong knowledge of foundation models, RAG, embeddings, vector search, tool use.
• Experience with cloud AI services, API/event integration, identity, networking, containers.
• Working knowledge of responsible AI, data governance, model risk and human oversight.
• Strong communication, facilitation and stakeholder-management skills.

Responsibilities

• Lead end-to-end architecture and delivery assurance of AI solutions.
• Translate business needs into AI use cases and acceptance criteria.
• Design models, data pipelines, embeddings, retrieval, orchestration and APIs.
• Evaluate build vs buy vs hosted options for safety, latency and cost.
• Validate designs with prototypes, tests and release thresholds.
• Guide MLOps/LLMOps, versioning, deployment and observability.
• Coordinate with cross-functional teams and align with enterprise standards.

Skills

AI architecture

MLOps

RAG & vector search

Prompt orchestration

Cloud AI services

Security & privacy

Stakeholder management

Model governance

Education

BS in CS/Engineering or related field

Tools

Azure AI / ML / OpenAI

APIs & microservices

Kubernetes / containers

Model hosting & deployment

Job description

Lead the end-to-end architecture and delivery assurance of secure, responsible, scalable AI solutions—from use-case discovery and experimentation to production monitoring and retirement. Partner with business, product, data science, engineering, operations, security, privacy, legal, risk and vendor teams to deliver measurable value through hands-on technical validation and sound architecture decisions.

Key Responsibilities

Translate business needs into AI use cases, feasibility assessments, measurable outcomes and acceptance criteria. Select predictive ML, generative AI, retrieval-augmented generation (RAG), agents, intelligent document processing or non-AI alternatives.

Design models, prompts, data and knowledge pipelines, embeddings, vector/hybrid search, orchestration, applications, APIs, cloud and identity. Define access-aware retrieval, guardrails, fallback behaviour and human oversight.

Compare build, buy and hosted, open-weight, customized or embedded AI options for quality, safety, data residency, licensing, latency, cost, control, vendor viability and lock-in.

Validate designs through prototypes, prompt/retrieval tests, model evaluations and implementation reviews. Set release thresholds for accuracy, relevance, groundedness, hallucination, toxicity, robustness, latency and cost.

Assess data fitness, provenance, consent, lineage and authorized use; address model risk, security threats, privacy, fairness, explainability, intellectual property and regulatory obligations.

Guide MLOps/LLMOps, versioning, reproducibility, deployment, rollback and operational readiness. Monitor quality, drift, safety, usage and cost throughout the lifecycle.

Maintain architecture decisions, intended use, limitations, risks, dependencies and review evidence. Align with enterprise standards and roadmaps, support required independent approvals, and assure implementation at lifecycle gates.

Communicate trade-offs, facilitate workshops, mentor teams and develop reusable patterns and controls. Coordinate shared capabilities while respecting business, model, data, platform and control-owner accountabilities.

Requirements

Required Qualifications

Typically 10+ years of relevant technology experience, with significant architecture responsibility and demonstrated delivery of production AI/ML solutions.

End-to-end architecture experience spanning models, data, retrieval, applications, integration, cloud, security and operations.

Strong knowledge of foundation models, RAG, embeddings, vector search, prompt orchestration, tool use and agents; practical understanding of classical ML, feature/data pipelines, evaluation and model failure modes.

Experience with cloud AI services, model hosting, API/event integration, identity, networking, secrets, containers and scalable deployment; MLOps/LLMOps, observability and model monitoring.

Working knowledge of responsible AI, AI security, privacy engineering, data governance, model risk and human oversight; ability to assess vendor documentation, prompts, evaluation evidence and deployment configurations.

Strong communication, facilitation and stakeholder-management skills; degree in computer science, engineering, data science or a related field, or equivalent experience.

Preferred Qualifications

AI delivery experience in financial services, credit unions or other regulated industries.

Azure AI, Azure Machine Learning, Azure OpenAI or comparable platforms; API management and infrastructure as code.

Familiarity with model gateways, AI safety tools, vector databases, knowledge graphs, feature stores, model registries and AI observability; commercial copilots, SaaS and open-source AI evaluation.

Enterprise architecture methods, recognized AI risk/security/governance frameworks, and relevant cloud, AI/ML, data, security or architecture certifications.