Skip to main content
Portfolio
HomeProjectsAboutCertificationsContact

© 2026 Portfolio. All rights reserved.

ContactAdmin

Portfolio

Pablo Navarro Zepeda

AI Software Engineer. Building systems that scale.

View projectsGet in touch
Download CV
10+years
60%latency reduction
2M+MAU (shipped product)

Work

Featured projects

View all

RAG pipeline at scale

Reduced retrieval latency by 60% and cost by 40% for a 10M-document knowledge base using hybrid search and custom reranking.

RAGLLM

Credentials

Certifications

View all

Apple Developer Academy

Apple

View certificate

Kind words

What people say

“

“Alex shipped our RAG stack in half the time we estimated. Clear communication, strong systems thinking.”

J
Jordan Lee

VP Engineering · Scale AI

“

“The observability work alone saved us weeks of incident response. Would work with again in a heartbeat.”

S
Sam Rivera

Head of ML · Fintech Co

Updates

Latest activity

  • Why RAG is still the best first step for enterprise LLM

    Short post on retrieval quality and when to reach for agents.

    Feb 6, 2026
  • Lessons from shipping on-device LLM to 2M users

    Latency, battery, and model selection tradeoffs.

    Feb 3, 2026

LLM observability platform

Built an internal platform for tracing, logging, and cost attribution across 50+ model endpoints. Cut debugging time by 70%.

LLMInfra

iOS + on-device AI assistant

Shipped an on-device LLM experience for a productivity app. Optimized for latency and battery; 2M+ monthly active users.

LLMiOS