AI Architect

4 DAYS ONSITE

Senior AI Architect should have the following skills sets/experience.

Experience Level: 8–10+ years in software engineering or data architecture, with at least 4+ years specifically dedicated to AI/ML systems design and deployment in production environments.

Cost Management: Proven ability to manage and optimize the computing costs associated with running heavy AI workloads (e.g., token optimization, GPU allocation).

Communication & Leadership: Exceptional ability to translate complex AI capabilities and limitations to C-suite executives and non-technical stakeholders.

Database Knowledge: Familiarity with SQL and data warehousing concepts (e.g., Snowflake, BigQuery) for data pipeline orchestration.

Essential Job Functions & Required Skills:

Enterprise AI Strategy: Lead the transition of AI projects from localized Proof of Concepts (PoCs) to scalable, enterprise-wide production systems.

Technology Selection: Evaluate and select the appropriate AI technologies (e.g., Deep Learning, Natural Language Processing, Computer Vision, Generative AI) based on business requirements, cost, latency, and data privacy constraints.

Architecture Design: Architect end-to-end AI pipelines, including data ingestion, model training/fine-tuning, deployment, and monitoring.

MLOps & Infrastructure: Design and implement robust MLOps practices for continuous integration, continuous deployment (CI/CD), and continuous training (CT) of AI models to prevent model drift and degradation.

Data & Integration: Work closely with data engineers to design the data architecture required for advanced AI, including vector databases and complex RAG workflows.

AI Governance & Ethics: Establish guardrails for AI usage, ensuring models are fair, transparent, secure against adversarial attacks, and compliant with data privacy regulations (e.g., GDPR, CCPA).

The table below lists the role(s) required for this engagement:

The desired country and location for work to be performed is USA. Candidate should come to office 4 days/week.

DELIVERABLES AND ACCEPTANCE CRITERIA

The resource will be engaged for the following deliverables:

Activity

Deliverables

Acceptance Criteria

1

AI Architecture Blueprint

A comprehensive AI Architecture Blueprint detailing how different AI models will integrate with existing enterprise applications and infrastructure.

Sign-off by the ITS Project Manager/ Technical Lead

2

Framework design

Standardized MLOps Frameworks and deployment pipelines for the data science and engineering teams to follow.

Sign-off by the ITS Project Manager/ Technical Lead

3

Delivery

Delivery of a secure Retrieval-Augmented Generation (RAG) Architecture or custom-tuned models tailored to internal company data.

Sign-off by the ITS Project Manager/ Technical Lead

4

Documentation

A formal AI Governance Document outlining security protocols, prompt injection defenses, and bias mitigation strategies.

Sign-off by the ITS Project Manager/ Technical Lea