In my independent client projects, I build and test AI-enabled automations, including an order-processing webhook pipeline with authentication, validation, duplicate checks, and modular subworkflows. I’ve also processed datasets of 10,000+ records and validated retry and error-handling workflows.
I built a sales pipeline automation that captures leads from forms and email, enriches and scores them through APIs, logs them in HubSpot and Google Sheets, and alerts the sales team in Slack. Other projects include an LLM classifier and router, a company RAG assistant, and a customer-service agent with persistent memory and tool-calling.
My AI training and evaluation work includes 436+ evaluation tasks at a 4.86/5 average accuracy score, along with SFT data curation and rubric-based response evaluation. Earlier, I wrote and edited 300+ pieces of technical documentation and research content for Web3 and engineering teams.
