Excel → KDP → commerce → software → AI.
The short version of a much longer story. Each phase exists because the previous phase created a problem that required a new skill.
Learning Excel because the business needed it
The technical journey did not begin with programming. It began with basic spreadsheets, product information and the need to organize business data more efficiently.
Amazon KDP and the first digital workflows
Kindle Direct Publishing introduced metadata, keywords, categories, book information and Amazon as a digital distribution system before Gifts Loft.
Gifts Loft and hands-on ecommerce
The work moved into physical products: listings, pricing, packaging, inventory, customers and daily marketplace operations.
Manufacturing, sourcing and the supply side
Wall clocks, home décor, product development, manufacturing coordination and China/Yiwu sourcing expanded the perspective from storefront to factory and supply chain.
Amazon India at serious operating volume
Listings, FBA, PPC, inventory, account health and catalog operations grew into a business operating at roughly ₹30–40 lakh per month during major periods.
From one marketplace to 20+ international accounts
The operating model expanded across international seller accounts, bringing cross-border catalogs, advertising, compliance, account health and repeatable processes.
When manual product work stopped scaling
Tens of thousands of products changed the problem. Catalog architecture, taxonomy, feeds and bulk updates became systems problems rather than individual listing tasks.
Shopify, WordPress and direct-to-consumer businesses
Universal Reads, Amegweb and Efinito added storefront ownership, conversion UX, Merchant Center, international markets and greater control over the customer journey.
Advertising, Merchant Center and SEO
Google Ads, Performance Max, Shopping, Amazon PPC, product feeds and technical SEO became part of the same operating system as catalog and storefront work.
Spreadsheets became Apps Script and repeatable workflows
The same repetitive catalog problems that had pushed Excel further eventually pushed the work into scripts, API calls and automated data transformations.
Python turned internal processes into applications
Python, Streamlit, FastAPI and structured batch processing made it possible to build tools around real ecommerce workflows rather than repeatedly solving them by hand.
Learning the production layer underneath the applications
Ubuntu, Docker, Coolify, PM2, PostgreSQL, Cloudflare and deployment troubleshooting became necessary as more systems moved beyond hosted storefronts.
Gergeant and the move from commerce tools to software products
Building a growing utility platform connected product engineering, SEO, browser workflows, backend services and production deployment into one public software venture.
Adding language models to the operating layer
OpenAI APIs and LLM workflows expanded automation from deterministic transformations into classification, enrichment, verification, content generation and structured decisions.
Agents, tools and AI × commerce
The current direction connects agents, APIs, Python services, n8n, business tools and human approval into systems that can carry more of the operating workload.