AI/ML Engineer with a strong backend foundation, building production-ready LLM systems using Gemini, Vertex AI, and Google ADK.

I focus on turning generative AI into reliable software — designing RAG pipelines, agent workflows with tool calling, and scalable APIs that connect models to real data and services. My backend background helps me ship AI systems with proper architecture: observability, latency control, cost awareness, and clean service boundaries.

• Retrieval-Augmented Generation (RAG) systems
• Agent workflows using Google ADK
• Tool/function calling pipelines with Gemini
• LLM-backed APIs (FastAPI / Python)
• Cloud-native deployments on GCP
• Evaluation + monitoring for LLM quality

AI/ML: Gemini, Vertex AI, Google ADK Backend: Python, FastAPI Cloud: GCP (Cloud Run, Cloud Functions, etc.) Data: PostgreSQL, Redis

LLMs are not features — they are systems. I build them with:

• deterministic components around probabilistic cores
• clear data boundaries
• measurable outputs
• production-grade deployment practices

What direction are you targeting?