# Visheshkumar Prajapati > AI Product Manager | Generative AI & Agentic Systems | Shipped LLM Platform Driving $3M+ Revenue | RAG, Multi-Agent AI & MLOps | Ex-Accenture Location: Greater Chicago Area, United States Profile: https://flows.cv/visheshkumar Most enterprise AI products fail not because the models are bad — but because no one connects what LLMs can do to what the business actually needs. That gap is where I operate — and it's how I drove $3M+ in incremental revenue and a 40% conversion lift across 50M+ annual customers at The Home Depot. As an AI Product Manager, I architect and ship enterprise-scale GenAI platforms and agentic AI systems. I spearheaded a multi-agent orchestration platform using GPT-4, Claude, and Gemini with RAG architecture and LangChain/LangGraph pipelines — cutting merchandising decision latency by 25% for 2,000+ store operations teams. I also built the personalized recommendation engine (collaborative filtering, embedding models, semantic search) that drove the $3M+ revenue impact. But revenue without trust doesn't scale. I instituted the company's responsible AI governance program — bias detection, SHAP explainability, human-in-the-loop validation, audit trails, and model rollback protocols — achieving 98% compliance and reducing rollback incidents by 60%. I designed a rigorous A/B experimentation framework (power analysis, 95% CI) across 20+ quarterly tests — validating $500K in annual inventory optimization. I also engineered reusable AI platform infrastructure across AWS SageMaker, Vertex AI, and Azure OpenAI — including LLM eval frameworks (RAGAS, BLEU), vector DB pipelines, drift detection, and CI/CD for ML — accelerating development velocity by 30% across 5 teams. Before Home Depot, I led product for an ML-powered Financial Portfolio Insights Platform at Accenture — 50,000+ users, embedded ESG scoring, predictive risk analytics, and GDPR/MiFID II compliance — improving client retention by 15%. I lead cross-functional teams of 12+, partner with C-suite on AI strategy, and think in systems — from inference costs to conversion funnels to regulatory audit trails. Core: GenAI Platforms | Agentic AI & Multi-Agent Orchestration | RAG Architecture | LLMs (GPT-4, Claude, Gemini) | LangChain & LangGraph | Recommendation Systems | LLM Eval (RAGAS, BLEU) | A/B Experimentation | MLOps | AI Governance | SHAP/LIME | HITL | Vector DBs | NLP | Prompt Engineering | Azure OpenAI & Copilot | AWS SageMaker & Bedrock | Vertex AI | GDPR, HIPAA, MiFID II MS Data Science — DePaul University | Microsoft Certified AI Product Manager Building AI that drives revenue and earns trust? Let's connect. ## Work Experience ### AI Product Manager @ The Home Depot Jan 2024 – Present | United States Leading AI product strategy and development for enterprise-scale generative AI and agentic systems — serving 50M+ annual customers across 2,300+ stores. Cross-functional leadership of 12 (engineering, data science, UX) with C-suite visibility on AI strategy and governance. 🚀 Platform & Agentic AI Architected and shipped enterprise multi-agent orchestration platform (GPT-4, Claude, Gemini) with RAG architecture and LangChain/LangGraph pipelines — cut merchandising decision latency by 25% for 2,000+ store operations teams Engineered reusable AI platform infrastructure: prompt templates, LLM eval frameworks (RAGAS, BLEU), vector DB pipelines, model deployment on AWS SageMaker & Vertex AI, drift detection, and CI/CD for ML — accelerated feature development velocity by 30% across 5 teams 💰 Revenue Impact & Personalization Spearheaded GenAI-powered personalized recommendation engine using collaborative filtering, embedding models, and semantic search — drove $3M+ incremental revenue and lifted conversion from 28% → 39% (+40%) across 50M+ annual customers in 6 months Validated $500K annual inventory optimization via ML-driven promotional pricing through rigorous A/B experimentation 📊 Data-Driven Experimentation Orchestrated A/B experimentation framework (power analysis, 95% confidence intervals) across 20+ quarterly tests — methodology adopted by 5 product teams Conducted user research with 200+ store managers and merchandisers — translated insights into prioritized roadmap delivering 40% improvement in task completion time Drove C-suite AI strategy visibility via 15+ KPI Power BI dashboards 🛡️ Responsible AI & Governance Instituted responsible AI governance program: bias detection, SHAP explainability dashboards, human-in-the-loop validation, audit trails, safety guardrails, and model rollback protocols Achieved 98% responsible AI compliance and reduced model rollback incidents by 60% ### Product Manager @ Accenture Jan 2020 – Jan 2023 | India Led end-to-end product development for ML-powered enterprise financial technology platform serving 50,000+ users, balancing regulatory compliance (GDPR, MiFID II) with AI innovation across embedded ML models and real-time analytics. 📈 Product Launch & Impact Launched enterprise Financial Portfolio Insights Platform with embedded ML models (ESG scoring, predictive risk analytics, asset optimization) to 50,000+ users Improved client retention by 15% through AI-driven automated ESG scoring, personalized risk analytics, and predictive optimization Designed Tableau dashboards visualizing portfolio performance, benchmark comparisons, and ESG metrics for real-time advisor decisions 🎯 Product Strategy & Roadmap Owned product roadmap balancing regulatory requirements (GDPR, MiFID II), technical feasibility, and user needs Drove feature prioritization through quantitative and qualitative research from 1,000+ customers using Qualtrics, SurveyMonkey, and usage analytics Defined backlog for risk scoring, asset allocation, and portfolio rebalancing AI tools ⚡ Operational Excellence Reduced release delays by 10% through proactive cross-functional risk management across data quality, compliance, and integrations Led post-launch AI model monitoring and rapid incident response — accelerated resolution by 35% and reduced P1 incidents by 40% Incorporated learnings from incidents into roadmap iterations for continuous improvement ## Education ### Master of Science - MS in Data Science DePaul University ## Contact & Social - LinkedIn: https://linkedin.com/in/visheshkumar-prajapati-oza --- Source: https://flows.cv/visheshkumar JSON Resume: https://flows.cv/visheshkumar/resume.json Last updated: 2026-04-17