Applied LLM

Period: 2023 – 2024
Scope: LLM Applications · RAG · Conversational AI
Tools: OpenAI · Google Gemini · ElevenLabs

Overview

Built real-world LLM-powered applications focusing on prompt engineering, conversational memory, retrieval-augmented generation (RAG), and multimodal interaction.

This role emphasized:

• designing LLM system architectures, not just prompts
• managing chat memory and context compression
• integrating speech, text, and document pipelines
• building deployable AI-powered applications

Key Contributions

• Designed persona-driven conversational systems with persistent memory
• Built RAG pipelines with source-grounded answers
• Integrated speech-to-text (STT) and text-to-speech (TTS)
• Developed structured output prompting pipelines
• Delivered multiple end-to-end AI applications (web & desktop)

Works

• Persona Chat System
• Resume Tailoring Engine
• Live Interview Copilot
• RAG Document Assistant
• LLM Evaluation Pipeline

Tech Focus

Prompt Engineering · RAG · Chat Memory · OpenAI API · Gemini API · Whisper · TTS

Impact

• Delivered production-ready LLM-powered applications
• Improved response quality using memory and retrieval strategies
• Enabled multimodal AI interactions (voice + text)
• Built foundation for agent-based AI systems