Description

I’m an AI Architect and Engineer passionate about designing intelligent systems that perform at the edge of what’s technically possible. Over the past few years, I’ve built production-grade LLM, RAG, and multi-agent systems that have reduced latency by over 80%, boosted operational efficiency by 40%+, and delivered measurable business impact across healthcare, logistics, and SaaS domains. My work focuses on secure, scalable AI infrastructure—ensuring innovation never compromises compliance. Beyond engineering, I thrive on bridging human creativity with machine intelligence to turn ideas into reality. I’m now seeking opportunities to architect next-generation AI platforms and lead teams that push the limits of intelligent automation.

At CycleSyncAI Labs, I engineered a low-latency conversational AI system that achieved sub-300ms response times—making it virtually indistinguishable from human interaction. This innovation reduced average customer wait times by 80% and powered a voice-driven SaaS product now handling thousands of conversations daily. Beyond engineering, I led the technical sales effort that helped secure our first $250K in pre-seed funding by building and presenting live AI demos to investors and enterprise clients. I also architected a fully self-hosted LLM infrastructure using Qwen3-Coder-480B-A35B-Instruct and LiteLLM, enabling secure multi-agent collaboration while keeping 100% of sensitive data on-prem. These combined efforts proved my ability to merge deep technical expertise with business strategy—turning cutting-edge AI into tangible market value.

Languages

English
Native or bilingual

Workplace preferences

Remote only

Primarily works remotely

CycleSyncAI Labs
Software Engineer
February 2024 - Today (2 years and 4 months)
Angeles, Pampanga, Philippines
• - Engineered a low-latency, conversational voice agent that mimicked human level responsiveness, achieving a sub-300ms response time and reducing average customer wait times by 80%.
• - Architected and deployed a secure, self-hosted LLM infrastructure leveraging Qwen3-Coder-480B-A35B-Instruct, LiteLLM, and Crush, enabling 50+ team members to utilize internal models while ensuring 100% of sensitive data never left company servers, mitigating compliance risk.
• - Designed and deployed the core AI model for a Lead Magnet Generation SaaS platform, training on 10,000+ samples and collaborating with marketing to boost conversion accuracy by 35%, while delivering JSON-to-HTML forms that cut campaign launch time from days to minutes.
• - Implemented a centralized logging and behavioral tracking system for all LLM interactions, improving model usage visibility and identifying 15% cost-saving opportunities through usage pattern analysis.
• - Designed and integrated a dual-memory system for AI agents using Redis, providing short-term contextual memory for active sessions and long-term user memory, which increased user satisfaction scores by 25%.
• - Developed an automated customer support agent capable of handling meeting scheduling and a dynamic RAG system that successfully resolved 85% of unique customer inquiries without human intervention, reducing support ticket volume by 40%.
LLM Voice AI Agent Livekit RAG Systems Redis Dual-Memory Systems
ShypV
AI Engineer
April 2025 - October 2025 (6 months)
United Arab Emirates
• - Developed "Logistics Chatess," an AI analytics platform that enabled non technical users to generate SQL queries and data visuals using plain English, eliminating the need for technical skills.
• - Architected a secure multi-tenancy system with Keycloak for 50+ clients, ensuring 100% data isolation and achieving zero security breaches.
• - Engineered a SQL Agent that translated natural language into database queries, autonomously handling 85% of data requests and reducing the analytics team's workload by 60%.
• - Improved SQL Agent accuracy by 40% by building a continuous learning pipeline with a curated repository of 1,000+ SQL samples and a user feedback loop.
• - Implemented a hybrid search system (keyword, semantic, self-querying) that successfully resolved over 90% of ambiguous user questions without requiring clarification.
• - Pioneered "Medical Chatess," an AI diagnostic assistant that integrated with EMRs to cut down initial patient review and hypothesis generation time by 30% for clinicians.
• - Built a HIPAA-compliant data infrastructure for 10+ clinics, securing sensitive PHI/PII with granular access controls and full audit trails.
• - Developed the AI analysis engine for generating diagnostic and treatment suggestions, achieving a 92% clinical relevance score as validated by 15+ medical experts.
• - Managed key relationships with medical professionals to translate expert feedback into model improvements, directly addressing real-world diagnostic challenges.
Exilion Technologies Inc
Software Engineer
January 2024 - February 2025 (1 year and 1 month)
Vaughan, ON, Canada
• - Executed a character-based transformer spelling correction model with Triton and ONNX, reducing inference latencies to below 178ms at 35 requests per second, improving response efficiency by 31.3%.
• - Conceptualized an AI-driven healthcare chatbot using a Large Language Model and Reinforcement Learning, resulting in a 25% increase in patient engagement and a 30% reduction in customer service response times.
• - Ensured HIPAA compliance, reducing data breaches by 28.5% while managing cloud-based databases for a healthcare organization.
• - Collaborated with data engineers and data scientists to architect scalable and efficient data pipelines for big data processing.
• - Implemented WebRTC solutions for real-time communication, contributing to innovative app features and enhancing user interaction.