Machine Learning / AI Engineer, Consultant building agentic AI systems, multimodal ML products, and production-grade data workflows.

I work at IBM on multi-agent systems, RAG, browser automation, and cross-cloud AI infrastructure. I also conduct medical AI research with Toronto General Hospital at UHN, where I build multimodal and deep learning systems for clinical prediction.

• Built a multi-agent Text-to-SQL system on AWS that reached 93% accuracy in 8 weeks and supported a $300K+ enterprise contract.
• Developed a multimodal agentic RAG system with 97% answer relevancy and cross-modal retrieval across text and image data.
• Built a cross-cloud agent interoperability platform connecting Azure Foundry, AWS Bedrock AgentCore, and LangGraph, reducing API calls by 73%.
• Researched and engineered OpenBrowser, an agentic browser framework for SAP automation that benchmarked at 3x cheaper and 20% faster through a DOM-first CDP approach.
• Engineered a transaction-level P&L rebuild pipeline in Python reconciling EBITDA across 15 entities (SAP, ECC, QuickBooks) with 67 rules; delivered BigQuery warehouse, LookML semantic layer, and Next.js AI copilot with Vertex AI Gemini.
• Deployed Gemma 4 31B on SageMaker (4x A10G) with 5-knowledge-base RAG on Bedrock, achieving 6x speedup (12 to 2 min) via ThreadPoolExecutor with WAF geo-lock to Canada.
• Built data workflows at Sanofi that processed billions of rows and reduced manual reporting cycles from days to seconds.

• Multi-agent systems, RAG, browser automation, and LLM application engineering
• Applied machine learning for healthcare, enterprise workflows, and multimodal data
• Full-stack AI products with React, TypeScript, React Native, Expo, Node.js, and AWS

• OpenBrowser AI: Agentic browser framework with CLI tooling, MCP server support, AWS deployment, and reinforcement learning pipelines.
• Job Seeker AI Agent: Multi-agent system that analyzes job postings and tailors resumes for stronger ATS alignment.
• Chat With Documents and Web: RAG application using LangChain, Chroma, and Hugging Face embeddings across documents and web sources.
• Chat With Image: Multimodal assistant built with IBM watsonx and Llama 3.2 Vision for image-aware conversations.

• Machine Learning / AI Engineer, Consultant, IBM Built multi-agent AI systems across AWS, multimodal RAG, cross-cloud interoperability, and agentic browser automation for enterprise use cases.
• Machine Learning Research Student, Toronto General Hospital / UHN Developed multimodal and deep learning systems for clinical prediction, including pathology-slide pipelines and survival prediction models across multi-center datasets.
• Data Scientist, Sanofi Built ETL and self-service analytics workflows on Python, SQL, Snowflake, and Streamlit, processing billions of rows for vaccine manufacturing operations.
• Full Stack Developer Intern, Leslie Dan Faculty of Pharmacy, University of Toronto Built an LLM-powered paper-screening agent and a full-stack platform with Node.js, React, Next.js, and TypeScript.

Georgia Institute of Technology
Master of Science in Computer Science, Specialization in Artificial Intelligence

University of Toronto
Honours Bachelor of Science with Co-op, with High Distinction
Data Science and Machine Learning Specialist (Co-op), Computer Science Major, Economics Minor
Accepted to 3 graduate programs: Fully Funded MSc/PhD at Institute of Medical Sciences (UofT), MSc Applied Computing AI Specialization at Department of Computer Science (UofT), and Master of Data Science and AI at University of Waterloo.

• Portfolio: billy-enrizky.github.io/portfolio
• CV: View Resume
• LinkedIn: linkedin.com/in/enrizky-brillian
• GitHub: github.com/billy-enrizky
• Email: billy.suharno@gmail.com

If you are building agentic AI, applied machine learning systems, or production AI products, feel free to connect.