{"canonicalUrl":"https://humetric.net/p/profile-2e72c0d049864732872e9ad0f0c516a7","profileVersion":"5dc69a50-9387-4b11-ac00-dc2ddb0b2afc","lastUpdated":"2026-08-20T02:52:17.39586+00:00","profile":{"display_name":"Anonymous profile","last_updated":"2026-08-20","slug":"profile-de760124858048c9","headline":"AI/ML Engineer","summary":"An AI/ML engineer with over 4 years of experience building production machine learning and generative AI systems in fintech and healthcare. Expertise spans the full ML lifecycle, including LLM fine-tuning, RAG development, and deploying scalable agentic applications with robust data engineering foundations.","years_experience":4,"skills":["Agentic AI","Anthropic Claude","Apache Airflow","AWS","Azure","Bash","BentoML","dbt","Docker","Embeddings","ETL/ELT","FAISS","FastAPI","Feast","GCP Vertex AI","GitHub Actions","Hugging Face Transformers","Kafka","Kubernetes","LangChain","LangGraph","LangSmith","LightGBM","Llama 3","LlamaIndex","LoRA","MLflow","Model Context Protocol","NumPy","OpenAI GPT-4o","pandas","PEFT","pgvector","Pinecone","Prompt Engineering","PySpark","Python","PyTorch","QLoRA","RAG","RAGAS","scikit-learn","Semantic Search","sentence-transformers","Snowflake","Spark","SQL","TensorFlow","vLLM","Weaviate","Weights & Biases","XGBoost"],"experiences":[{"id":"380b14ba-bcba-4ee5-adc7-c038faa3ed06","person_id":"b3514172-2edb-4645-8c3c-79298f125567","title":"Data Scientist","company":"Saayam for All","start_date":"June 2025","end_date":"July 2026","description":"Built an LLM-powered request classification and routing service using GPT-4o and Llama 3 to automate aid request categorization and improve routing accuracy.","position":0},{"id":"d4b4b1b1-630a-44ab-b025-a4b36cd20849","person_id":"b3514172-2edb-4645-8c3c-79298f125567","title":"Data Scientist","company":"Hamilton Digital Assets","start_date":"October 2024","end_date":"May 2025","description":"Developed a gradient-boosted fraud detection model and an LLM-based alert-summarization tool to streamline analyst review processes.","position":1},{"id":"af49eadf-1931-4f1c-9ddf-8e08bc8a0885","person_id":"b3514172-2edb-4645-8c3c-79298f125567","title":"Data Engineer","company":"Mphasis","start_date":"August 2020","end_date":"July 2023","description":"Developed batch ingestion pipelines and automated reconciliation jobs to consolidate banking trade data and improve reporting efficiency.","position":2}],"educations":[{"id":"f9d7e9a9-2acf-48cd-995f-6f761bd6d880","person_id":"b3514172-2edb-4645-8c3c-79298f125567","school":"Illinois Institute of Technology","degree":"Master of Science","field":"Data Science","start_year":2023,"end_year":2025}],"projects":[{"id":"dcc1baea-ec99-497a-8534-5f308ccaa370","person_id":"b3514172-2edb-4645-8c3c-79298f125567","name":"Health Insurance Benefits & Coverage Verification Agent","description":"Built an agentic LLM assistant using LangGraph and RAG to explain insurance coverage and out-of-pocket costs for medical procedures with source citations.","url":null,"technologies":["Python","LangGraph","GPT-4o","pgvector","FastAPI","Docker"],"position":0},{"id":"83f40277-6fd7-4eba-91a4-1d2c23af69fd","person_id":"b3514172-2edb-4645-8c3c-79298f125567","name":"Real-Time Fraud Detection & Agentic Dispute Triage","description":"Developed a real-time fraud scoring system using Kafka and XGBoost, complemented by a LangGraph agent for automated dispute resolution drafting.","url":null,"technologies":["Python","Kafka","Feast","XGBoost","LangGraph","Redis","FastAPI","MLflow","Kubernetes","AWS"],"position":1}]},"chunksVersion":"profile-chunks-v3","approvedDiscoveryQueries":["AI/ML Engineer","Python PyTorch TensorFlow"],"evidence":[{"id":"identity:0","type":"identity","ordinal":0,"locale":"en","text":"AI/ML Engineer. 4 years experience","fieldKeys":["headline","years_experience"]},{"id":"summary:0","type":"summary","ordinal":0,"locale":"zh","text":"AI/ML Engineer. Summary / 簡介: An AI/ML engineer with over 4 years of experience building production machine learning and generative AI systems in fintech and healthcare. Expertise spans the full ML lifecycle, including LLM fine-tuning, RAG development, and deploying scalable agentic applications with robust data engineering foundations.","fieldKeys":["summary"]},{"id":"skills:0","type":"skills","ordinal":0,"locale":"zh","text":"AI/ML Engineer. Skills / 技能: Agentic AI, Anthropic Claude, Apache Airflow, AWS, Azure, Bash, BentoML, dbt, Docker, Embeddings, ETL/ELT, FAISS, FastAPI, Feast, GCP Vertex AI, GitHub Actions, Hugging Face Transformers, Kafka, Kubernetes, LangChain, LangGraph, LangSmith, LightGBM, Llama 3, LlamaIndex, LoRA, MLflow, Model Context Protocol, NumPy, OpenAI GPT-4o, pandas, PEFT, pgvector, Pinecone, Prompt Engineering, PySpark, Python, PyTorch, QLoRA, RAG, RAGAS, scikit-learn, Semantic Search, sentence-transformers, Snowflake, Spark, SQL, TensorFlow, vLLM, Weaviate, Weights & Biases, XGBoost","fieldKeys":["skills"]},{"id":"discovery:0","type":"discovery","ordinal":0,"locale":"zh","text":"AI/ML Engineer. Role / 職稱: AI/ML Engineer. Experience / 年資: 4 years experience. Expertise / 專長: Agentic AI, Anthropic Claude, Apache Airflow, AWS, Azure, Bash, BentoML, dbt, Docker, Embeddings, ETL/ELT, FAISS","fieldKeys":["headline","years_experience","skills"]},{"id":"experience:0","type":"experience","ordinal":0,"locale":"zh","text":"AI/ML Engineer. Experience / 經歷: Data Scientist at Saayam for All. June 2025 to July 2026. Built an LLM-powered request classification and routing service using GPT-4o and Llama 3 to automate aid request categorization and improve routing accuracy.","fieldKeys":["experiences"]},{"id":"experience:1","type":"experience","ordinal":1,"locale":"zh","text":"AI/ML Engineer. Experience / 經歷: Data Scientist at Hamilton Digital Assets. October 2024 to May 2025. Developed a gradient-boosted fraud detection model and an LLM-based alert-summarization tool to streamline analyst review processes.","fieldKeys":["experiences"]},{"id":"experience:2","type":"experience","ordinal":2,"locale":"zh","text":"AI/ML Engineer. Experience / 經歷: Data Engineer at Mphasis. August 2020 to July 2023. Developed batch ingestion pipelines and automated reconciliation jobs to consolidate banking trade data and improve reporting efficiency.","fieldKeys":["experiences"]},{"id":"education:0","type":"education","ordinal":0,"locale":"zh","text":"AI/ML Engineer. Education / 學歷: Master of Science. Data Science. Illinois Institute of Technology","fieldKeys":["educations"]},{"id":"project:0","type":"project","ordinal":0,"locale":"zh","text":"AI/ML Engineer. Project / 專案: Health Insurance Benefits & Coverage Verification Agent. Built with / 使用技術: Python, LangGraph, GPT-4o, pgvector, FastAPI, Docker. Built an agentic LLM assistant using LangGraph and RAG to explain insurance coverage and out-of-pocket costs for medical procedures with source citations.","fieldKeys":["projects"]},{"id":"project:1","type":"project","ordinal":1,"locale":"zh","text":"AI/ML Engineer. Project / 專案: Real-Time Fraud Detection & Agentic Dispute Triage. Built with / 使用技術: Python, Kafka, Feast, XGBoost, LangGraph, Redis, FastAPI, MLflow, Kubernetes, AWS. Developed a real-time fraud scoring system using Kafka and XGBoost, complemented by a LangGraph agent for automated dispute resolution drafting.","fieldKeys":["projects"]},{"id":"discovery:1","type":"discovery","ordinal":1,"locale":"zh","text":"AI/ML Engineer. Approved discovery query / 核准搜尋語句: AI/ML Engineer","fieldKeys":["headline","desired_queries"]},{"id":"discovery:2","type":"discovery","ordinal":2,"locale":"zh","text":"AI/ML Engineer. Approved discovery query / 核准搜尋語句: Python PyTorch TensorFlow","fieldKeys":["headline","desired_queries"]}]}