Agentic AI Future by Seena Ganesh (Staples)

Agentic AI Future: Agents Reshaping Digital Transformation and API Strategy
Seena Ganesh, Vice President Engineering - B2C & B2B eCommerce & Digital AI at Staples

apidays New York 2025
API Management for Surfing the Next Innovation Waves: GenAI and Open Banking
Convene 360 Madison, New York
May 14 & 15, 2025

------

Check out our conferences at https://www.apidays.global/

Do you want to sponsor or talk at one of our conferences?
https://apidays.typeform.com/to/ILJeAaV8

Learn more on APIscene, the global media made by the community for the community:
https://www.apiscene.io

Explore the API ecosystem with the API Landscape:
https://apilandscape.apiscene.io/

Transcript

1. Let’s Get Started Presented By: Seena Ganesh VP, eCommerce Engineering & Digital AI A G E N T S R E S H A P I N G D I G I T A L T R A N S F O R M A T I O N A N D A P I S T R A T E G Y AGENTIC AI FUTURE
2. API ENDPOINTS TO INTELLIGENT AGENTS 1 2 3 4 5 Early 2000s: Static and Mechanical • Age of static endpoints SOAP, WSDL • Rigid contracts and mechanical • Know exactly what to ask for 2005-2015: API Products • The rise of REST and JSON • APIs became products • Developer platforms emerged 2015-2020: Open APIs • Explosion of APIs, SaaS • Gateways, Open APIs • Open banking, Headless 2020-2023: Cognitive AI • AI joins the party • Natural language • API based LLMs, GenAI 2023 – 2024: Agentic AI • Time of agents • Tell it what you want don’t tell how • Smart and intelligent
3. What are Agents? Agents are “Smart Interns” Task : “Book my flight to New York, find a hotel, and prepare a briefing for the meeting.’” The intern: 1. Figures out tools to use: Google, Expedia, Docs 2. Decides cheapest vs. best reviewed 3. Go through docs and prepare briefing 4. Knows my interests and decides for me It is an AI intern who never sleeps… Think & Plan Act Observe Output
4. Architectural Shift to Agent-Oriented “Red” “Women” “Formal” Search query with filters Traditional API Architecture User History Analyzer Agent “Find a red semi-formal dress for her which is more relaxed yet a good one of evening cocktail party. She is between 25-30 years and usually buys semi formal wear. She dislikes very flashy dresses” Text Processor Service API Search Engine Service API Search Application Service API Agentic AI Architecture Agent Orchestration Fashion Trends Agent Query Optimizer Agent “Red formal dress for Cocktail dinner” “She is between 25-30 years. buys semi-formals” AI Based Search Engine Color = “Red” Gender = “Women” Type – “Formal” “Current trend for women is relaxed fit" Semantic Analyzer Agent Task: “Find a red formal dress for a woman”
5. Retrieval Augmented Generation (RAG) Retrieval System Data Sources Orchestrator Large Language Model User Interface User Query Answer Augments a language model with a retrieval system that fetches data from an external knowledge base and enriches that to generating a response.
6. AI Search Application Front End Search Agents Orchestration Commercial Models Open-source Models Internal Agents Commercial Agents APIs User Data Buying History Previous search Terms Query Analyzer History Analyzer Trend Analyzer RAGs AI Models Data Sources Agents & APIs APIs External Agents Internal Models
7. Agents, APIs & Chains
8. Product Search Semantic and contextual search capabilities Conversational AI Customer engagement through conversational AI Pricing & Promotions Dynamic pricing and relevant promotions Recommendations Timely Product recommendations Dynamic UI UI based on contextual information Business Operations Automation of business processes 1 2 3 4 5 6 Omnichannel Commerce Use cases for Agents
9. Steps to Start a Successful Business Evolution of API Strategy in the Agentic Era Customer Focus Design Orientation Documentation Structure Security Discovery Developer Data Oriented Human Readable Rigid & Mechanical Application level Prescriptive API Strategy & Agents Intent Oriented Agents & LLMs Intuitive & Situation Behavior Level Dynamic Agentic AI Strategy
10. The Customer Journey in Marketing Leadership Strategy for Agentic AI Objectives & KPIs Value Data Culture Process Enterprise Data Organizational Transformation Iterative Development
11. 1 2 3 4 5 6 Recommendations Human-first AI Move beyond productivity and consider use cases to fully deliver AI promise. Launch MVPs Develop MVPs with multiple AI techniques to adapt to learning curve. Best Practices Enhance the success rate of AI initiatives by applying AI engineering practices. Literacy Foster the responsible use of AI through AI literacy training program. Security & Governance Adopt security & governance practices for AI regulation and risk management. Cognitive Diversity Aim for AI based problem solving w/o compromising cognitive diversity.
12. Closing Thoughts Agents & APIs Agents and APIs will play a key role in AI enablement Opportunity Transformative opportunity with legal and ethical risks Solutions Agents, LLMs, LAMs, RAGs, Langchains Concerns Security, identity, privacy, legal, ethical, compliance Caution! Low-risk use cases first and mature governance Transformation Transform the org for a new way of autonomous execution 1 2 3 4 5 6
13. Questions? https://www.linkedin.com/in/seenaganesh