Joseph I.

I build LLM apps and AI automation that turn manual work into usable products: AI chat interfaces, resume and document tools, and content workflows with human approval. My stack includes Python, FastAPI, React, Anthropic Claude, and Google Gemini. If you need to add AI to an existing product or build an AI-powered MVP, I can help with: • LLM API integration: connect Claude or Gemini to your application and design the workflow around your users. • AI chat and document generation: conversational interfaces, tailored content, and custom DOCX/PDF exports. • Workflow automation: content generation with review and approval steps before publication. • Python APIs and deployment: FastAPI backends, Docker, and integration with your frontend. • Applied machine learning: classification, computer vision, model evaluation, threshold tuning, and fairness audits. Selected projects ResumeFit — Built and shipped a production resume-tailoring app using React and the Anthropic API. Built a second resume application with an AI career-advisor chat and a custom DOCX/PDF generation engine. AI content automation — Built a Gemini-powered content platform with a mandatory human-approval step before publishing. FruitNet — Trained and deployed a PyTorch Faster R-CNN object-detection API with an analytics dashboard, live on Hugging Face Spaces. Vertex AI classification — Trained and deployed churn and credit-risk models on Google Cloud Vertex AI, including model comparisons and a fairness audit. My background in electrical/electronic engineering and telecom field engineering shapes how I work: check data quality early, evaluate outputs, and build systems that can be maintained. Tools: Python, FastAPI, React, SQL, Anthropic API, Google Gemini, Docker, Google Cloud Vertex AI, PyTorch, scikit-learn, Power BI, and Tableau. Tell me what you want users to accomplish, the tools you already use, and your target timeline. I can help turn that into a clear build plan and a working application.