Meet Patibandha

• Architecting AI-powered Master Data Management (MDM) platform on Google Cloud Platform,
serving as single source of truth for restaurant, product, and transaction data across North
America and UK operations.
• Designing strategic AI infrastructure roadmaps integrating AI-driven automation across cloud
compute, database, storage, and network layers.
• Deploying Vertex AI models (NLP, forecasting) for intelligent data classification, predictive
infrastructure capacity planning, and operational anomaly detection with real-time alerting.
• Managing Terraform-based IaC including VPC networking, firewall rules, cross-project IAM, and
enterprise-level infrastructure automation standards.
• Built custom BDRLogger framework for structured logging to Cloud Logging and Dynatrace,
adopted across 4 environments (dev, test, stg, prd).
• Impact: Built enterprise-grade reusable microservice libraries adopted organization-wide; led
GCP cost optimization delivering significant hard-dollar savings.

• Designed AI/ML prediction models and data pipelines using PyTorch, TensorFlow, and Ray
Cluster; built LLM models with VectorDB and GraphDB for intelligent enterprise knowledge
retrieval.
• Championed Azure APIM / AI Gateway governance program across enterprise LLM deployments
— production guardrails, PII/PCI scan violation testing, OAuth2 client enforcement (EASE-ALN,
EASE-SmartGen, gw_pii, gw_non_pii), and monitoring for regulated GenAI endpoints.
• Drove AIOps adoption across compute, database, storage, and network layers; delivered
predictive monitoring and self-healing capabilities reducing infrastructure incidents.
• Leveraged AWS SageMaker and Bedrock for end-to-end ML lifecycle; automated AI training
environments using Terraform across server, compute, storage, and networking layers.
• Impact: Enhanced ML model accuracy through systematic fine-tuning; reduced data processing
time by 30%.

• Architected global Data Science workbench on AWS with automated compute, storage, and
backup provisioning; integrated Terraform into CI/CD for infrastructure automation including
database deployment and network security.
• Established coding standards and TDD practices; mentored teams on deep learning, MLOps, and
infrastructure automation best practices across multiple therapeutic research groups.
• Impact: Implemented CI/CD pipelines reducing deployment times by 40%; pioneered deep
learning models achieving high predictive accuracy in therapeutic research.

• Developed GCP-based image factory with automated pipelines; built Kafka streaming pipelines
sourcing from Oracle, IBM DB2, and Spanner DB.

• Built CI/CD pipelines and automated testing for Digital Commons Network; led Scrum
ceremonies; delivered Python/Django/Docker/AWS applications with 25% improvement in
release cycles.

• Developed ETL APIs and data mining workflows; leveraged PySpark with Hadoop for large-scale
data processing and automated pipeline orchestration.