Open to opportunities

Kevin Wang

@kevinwang11

Message

Senior AI/ML Engineer building production LLM, RAG, and decisioning systems at scale.

United States

Message

What I'm looking for

I’m looking to lead production GenAI/ML platforms—RAG, agents, ranking, and MLOps—where I can ship measurable improvements with strong engineering rigor and cross-team collaboration.

I’m a Senior AI/ML Engineer with ~10 years of experience designing and deploying large-scale machine-learning and AI systems, including LLM applications. I build production-grade AI platforms that combine ML, LLM orchestration, retrieval systems, and real-time decisioning at scale.

At Orita, I architected an enterprise AI platform integrating ML models, LLMs, and retrieval systems for decisioning and workflow automation across product and data teams. I built enterprise RAG systems with LangChain, LangGraph, and LlamaIndex using Pinecone, pgvector, and hybrid retrieval (vector + BM25), improving retrieval relevance by 25–40%. I also developed multi-agent AI workflows for autonomous task execution and tool-based reasoning, and I integrated OpenAI and Anthropic Claude APIs into production surfaces.

I’ve also delivered end-to-end ML and AI capabilities across teams at Pacaso and Robinhood—recommendations, semantic search, GenAI buyer assistants (RAG over structured sources), fraud detection, churn prediction, feature stores, monitoring, and ML governance. My approach emphasizes evaluation and observability (hallucination rate, latency, output quality), robust MLOps (drift detection, structured logging, CI/CD), and scalable backend services with FastAPI, Kubernetes, and event-driven architectures.

Experience

Work history, roles, and key accomplishments

Current

Senior AI Engineer

Current

Orita

Sep 2024 - Present (1 year 9 months)

Architected an enterprise AI platform integrating ML, LLMs, and retrieval systems for decisioning and workflow automation across multiple product and data teams. Built enterprise RAG with LangChain/LangGraph/LlamaIndex and Pinecone/pgvector using hybrid retrieval, improving retrieval relevance by 25–40%, and implemented multi-agent workflows plus LLM evaluation and observability.

RAG Systems LLMs (LangChain LangGraph LlamaIndex Pinecone Pgvector Multi Agent Systems LLM Evaluation

AI Engineer

Pacaso

Feb 2023 - Jul 2024 (1 year 5 months)

Built learning-to-rank recommendation systems and semantic search using embeddings with Pinecone and Weaviate, improving retrieval accuracy and relevance. Developed GenAI buyer assistance and document intelligence using RAG over property data/contracts plus OCR and LLM-based extraction, and delivered pricing optimization with continuous training, drift monitoring, and Airflow ML pipelines deployed

PyTorch XGBoost Pinecone Weaviate RAG Systems Airflow AWS Drift Monitoring

AI Engineer

Robinhood

Jan 2020 - Jan 2023 (3 years)

Developed real-time fraud detection using gradient boosting, anomaly detection, and streaming event pipelines. Built churn prediction and personalization ranking models with centralized feature storage to reduce train/serving skew, and implemented production monitoring for drift/latency plus ML governance and A/B testing to support model lifecycle management.

Real Time Fraud Gradient Boosting Streaming Pipelines Churn Modeling A B Testing Monitoring For Drift

Software Engineer

Robinhood

Apr 2019 - Jan 2020 (9 months)

Built high-throughput trading and portfolio systems using Python, Node.js, and microservices architecture. Designed event-driven pipelines for real-time financial data processing and ML feature generation, implemented identity verification and KYC/AML workflows, and optimized backend latency, scalability, and reliability under high traffic.

Python Microservices REST APIs fastAPI Scalability

Software Engineer

Redfin

Oct 2017 - Feb 2019 (1 year 4 months)

Built large-scale real estate search and ranking systems using user behavior and MLS data. Developed geospatial search and map-based filtering, improved ranking via ML feature pipelines, and increased conversion and relevance through A/B testing while enhancing backend performance with caching and query optimization.

Geospatial MLS Data Feature Engineering Recommendation Systems A B Testing Caching Query Optimization

Software Engineer Intern

Google

Jun 2017 - Sep 2017 (3 months)

Built internal tools and dashboards for distributed systems monitoring and debugging. Improved performance of internal systems through query optimization and contributed to production engineering tooling used across multiple teams.

Query Optimization Internal Tooling