Hire Paan D. - Generative AI Architect

* Cross-team collaboration to identify AI use cases that enhance productivity and deliver improved ROI. * Google ADK with Vertex AI Implementation: Solved four different Business use cases * Led agentic workflows utilizing Google ADK Framework/Gemini to aid the SRE team in prompt issue resolution and troubleshooting. Design/ Developed a multi-agent workflow utilizing Google ADK. Enabled A2A protocol. * Led/ Built and deployed Retrieval-Augmented Generation applications, reducing research time by 40% and enabling prompt data-driven decisions for the clients. * Led/ Built the MCP server for claim & policy summary. Build a Knowledge graph using Spanner. * Led, designed, and delivered an end-to-end AI autonomous agent (Iris) for the Marketing team (This is a multi-agent system that scores emails, fax, and mail content and suggests improvements) using MS Copilot Studio, Power Automate workflows and triggers, SharePoint, and Power BI. * Designed and delivered an end-to-end AI autonomous agent for Image suggestions and brand guidelines. * Production support for LLM agents (LLM Ops).

* Designed and implemented an enterprise-grade Data Platform architecture from scratch, including * technology and tool selection. * Created a visionary Data and AI digital strategy blueprint for Good Food Holding Inc., positioning it for * future scalability and innovation. * Engineered a custom metadata-driven ingestion framework, streamlining data integration and * processing workflows.

* Technical Enablement: * · Transforming Five Below IT from vendor driven/ legacy systems to more scalable cloud platform including Azure, Amazon, Oracle and Google cloud * · Enabling and building a new, modern Enterprise Data Platform, Customer Data Platform, and Analytics infrastructure from the ground up. * · Developed various pattern-based, metadata-driven, reusable frameworks leveraging Databricks/Spark's distributed computing for scalability and growth. This includes near-real-time and batch data ingestion, as well as auditing. * · Enabled a homegrown Data Quality (DQ) Framework for FIVE BELOW to assess data accuracy, validity, consistency, and timeliness, which will help save on licensing costs in the long run. * · Enabled a microservice-based ingestion framework capable of processing large volumes of data. * · Developed an Integration Bus – a plug-and-play Enterprise Service Bus (ESB) on top of a Kafka-based events-queuing infrastructure with Spark-Databricks compute, utilizing MongoDB, Delta-Databricks, and ADLS as persistent stores for use by all product groups within the company. * · Introduced delta-sharing/Lakehouse capabilities into the ecosystem, which eliminates duplicate data persistence. * · Enabled a cost-effective, real-time Kafka (HDFS-based) streaming framework that various FIVE BELOW product teams can use to stream data from HDFS sources, including Kafka and BLOB storage. * · Optimized Databricks data management techniques, including vacuuming and Z-ordering, to speed up query response times. * Business Enablement: * Modern IT platforms have enabled us to automate many of our manual tasks, freeing up our employees to focus on more strategic work. * Our new supply chain management system has helped us to reduce our inventory costs. * We have implemented a new customer relationship management (CRM) system that has helped us to improve our customer service levels and increase sales. * We are using artificial intelligence (AI) to personalize our marketing campaigns.