Kuldeep User

I'm looking to build and scale practical AI products, especially agent driven systems that solve real business problems through data, automation, and intelligent decision making. At PricewaterhouseCoopers Services LLP, I built a natural-language interface that let business teams query CRM data without SQL. My planner agent generated, executed, and validated SQL step by step, achieving 94% first-attempt accuracy across 100 queries on 10 databases. I've built autonomous AI products including ContentFactory, which produced 88 narrated cinematic videos through an LLM-agent workflow, and ValueAI, a multi-agent system that found 10+ verified underpriced e-commerce deals daily. I also developed an AI interview learning platform using QLoRA fine-tuned Mistral-7B and Faster Whisper, maintaining under-two-second response latency across 50+ mock interviews. I'm currently studying sales operations, incentive structures, broker engagement, and compensation models at PropertyPistol, combining data analysis with scalable business strategy.