● Agent Memory & Systems · Duke ECE PhD (2026)
ML Engineer / Researcher — Agent Memory · Retrieval · Production Systems
I build the write/retrieve/forget lifecycle for agent memory. SAGE — a novelty gate for efficient memory evolution — beats Mem0 7/7 on the LoCoMo benchmark at ~3.4× lower cost and ~2.5× lower latency. Public arXiv preprint + single-command reproducible code.
Open to
Recommendation & retrieval, LLM/VLM reliability, or agent memory — I bring production experience and research depth to all three.
Candidate generation, embedding retrieval, and ranking at production scale.
Pinterest GraphSAGE + Faiss · 1% revenue uplift · BERT broad-match CTR
Hallucination detection, uncertainty, and trust & safety for foundation models.
TMLR 2026 cross-modal consistency · GPT-4V, Qwen-VL, LLaMA-VL benchmarks
Memory infrastructure for LLM agents across the write / retrieve / forget lifecycle.
SAGE — beats Mem0 7/7 · 3.4× cheaper · code public
The thesis
Agent memory is what holds my work together — the same machinery I've shipped and researched for years, wearing an agent costume.
A memory system embeds, stores, and retrieves top-k under latency and cost constraints, then ranks by relevance to context. That's a two-tower / ANN problem — exactly the Pinterest GraphSAGE + Faiss systems I shipped to production.
Pinterest · GraphSAGE + Faiss · 1% revenue uplift
What to write, consolidate, and forget — and how to avoid catastrophic interference with old knowledge — is the continual-learning problem stated verbatim. My Samsung work (continual + federated, 3 patents) is the backbone of the write path.
Samsung · continual + federated · 3 patents
Memory at production scale means latency budgets, cost per query, and reliable serving. SAGE's cost and latency reductions are systems results — and my reliability research (TMLR 2026) keeps the read/write paths trustworthy.
SAGE · cost + latency wins · TMLR 2026 reliability
How I got here
Research Fellow / Deep Learning Research Intern · Samsung Semiconductor · SOC R&D Lab
Research Intern — Ads Retrieval & Targeting · Pinterest Labs
PhD Research — Duke University · Advisor: Prof. Ricardo Henao
Memory Management System for AI Agents · Duke University · the convergence
Selected work
Agent Memory · ARR — under review · code public
A novelty gate for efficient memory evolution in agentic LLMs. Frames memory evolution as novelty detection via density estimation, so the system writes/consolidates only what matters.
Continual Learning · Samsung · 3 patents
Communication-efficient federated learning, sustainable continual learning, and continual few-shot learning — the write-path backbone of agent memory.
Retrieval / RecSys · Shipped to production
Graph-based advertiser-similarity retrieval (GraphSAGE + Faiss ANN) plus a multitask BERT broad-match model, integrated into Pinterest's Spinner workflow.
Trust & Safety · TMLR 2026
A cross-modal consistency framework that detects hallucinations in vision-language models by comparing visual- and text-grounded reasoning paths.
Toolkit
SAGE: A Novelty Gate for Efficient Memory Evolution in Agentic LLMs
ARR (under review)
Fallback-Enabled Closed-Set Classification: Cross-Modal Consistency in Vision-Language Models
TMLR 2026
GAN Memory with No Forgetting
NeurIPS 2020
Model Recycling Framework for Multi-source Data-free Supervised Transfer Learning
IEEE MLSP 2025 (Oral)
Toward Sustainable Continual Learning: Detection and Knowledge Repurposing of Similar Tasks
IEEE MLSP 2025
A Holistic Approach to Interpretability in Financial Lending
Decision Support Systems 2022
Let's talk
Available June 2026. Production systems experience plus research depth across retrieval, continual learning, and reliability. Reach out.