Passed the Designing & Implementing Microsoft DevOps Solutions exam via PearsonVue — not just a course completion, but a proctored, high-stakes Microsoft certification exam. Covers Azure DevOps pipelines, CI/CD, infrastructure as code, monitoring, and security.

Worked at Scale AI for over a year evaluating outputs from frontier LLMs — GPT-4, Claude, Gemini — across code generation and reasoning tasks. Designed evaluation frameworks, benchmarked 4 frontier models, and developed failure taxonomies that shaped annotation guidelines for RLHF training pipelines.

Independently built and shipped 3+ full AI products live to production: HireOS (LangGraph + Gemini 2.0 Flash AI hiring assistant), KundaliAI (TanStack Start + Supabase Vedic astrology engine), and LaptopPulse (Python performance daemon). Full ownership — architecture, AI integration, UX/UI, payments, deployment.

Completed the full Microsoft AI Product Manager Professional Certificate on Coursera — covering product management for AI systems, market research, product strategy, UX/UI for AI, and go-to-market execution. A comprehensive, multi-course professional certification.

Completed Microsoft's AI Skills Challenge during AI Skills Fest 2026 — a global event bringing together millions of learners to build AI literacy and applied AI skills. Earned a verifiable credential for demonstrating AI application capabilities across Microsoft's ecosystem.

Built a 200+ piece art portfolio spanning AI-generated artwork (Stable Diffusion, Midjourney, ComfyUI) and traditional handmade pieces. The same systematic creativity applied to building agentic systems — turning complex prompts into coherent visual outputs. Available as prints in the Art Store.

Completed Bachelor of Technology in Computer Science & Engineering from Abdul Kalam Technical University (AKTU) in 2025. Built foundational expertise in algorithms, data structures, software engineering, and AI/ML that underpins all product development and LLM evaluation work.

Earned AMD AI Academy certification in GPU Optimization for LLM Inference — covering vLLM, ROCm, performance tuning for large language model serving, and efficient GPU utilization. Directly applicable to production AI system deployment and scaling.