Hire the Best Data Scientists
in Vietnam

I build production AI pipelines that turn messy documents into structured data. 8 years in Python and data engineering. I specialize in document processing systems — the kind where you throw in a stack of scanned PDFs and get clean, validated JSON out the other side. ## What I actually build ### LLM-Powered Extraction Pipelines Multi-agent systems using Claude, GPT-4, and GPT-4o with structured output enforcement (Instructor, Pydantic). I design domain-specific agents that classify documents, extract fields, validate data, and reconcile conflicts across sources. Not prompt-and-pray — schema-enforced, retry-aware, production-grade. ### OCR & PDF Processing Amazon Textract (async API), PyMuPDF, PDFPlumber, Camelot, Azure Document Intelligence. I handle the real-world mess — scanned documents with poor quality, mixed orientations, concatenated multi-document PDFs that need splitting and boundary detection. Table extraction, form parsing, layout-aware chunking. ### Serverless & Event-Driven Architecture AWS Lambda, SNS, SQS, Step Functions, EventBridge, ECS/Fargate. I build pipelines where S3 uploads trigger processing chains — fan-out to parallel extractors, dead-letter queues for failures, CloudWatch monitoring for quality metrics. ### Healthcare Document Systems Currently running a production pipeline that processes medical records daily — licenses, clinical notes, immunization records, provider credentialing documents. Full pipeline: PDF ingestion → OCR → document classification → LLM extraction → deduplication → normalization → structured output. Built with HIPAA considerations. ### Legal Document Systems Production pipeline for UK cost-assessment workflows — ingests multi-document bill bundles (PDFs, scans, Excel ledgers, Outlook .msg), classifies them, extracts disbursements, receipts, invoices, and party data, reconciles figures across sources with a 5-case matching algorithm, and generates court-compliant inter-partes narratives. Every extracted figure traces back to its source page for audit defensibility. ### Speech-to-Text Pipeline - AWS Transcribe Medical (streaming WebSocket + batch S3 jobs) with HIPAA-compliant configuration - Deepgram integration as an alternate ASR backend - Speaker diarization (up to 30 speakers), stereo channel separation, custom medical vocabularies, specialty-specific vocab lists (e.g. orthopedics) - Multi-threshold confidence analysis for dot-phrase and quick-text detection ### LLM Clinical Note Generation - Instructor + Pydantic schema-enforced structured outputs (Claude / GPT-4 / Gemini) - Prompt Studio: a template-authoring system where clinicians design reusable note templates with variables, segment bindings, and inline citations back to the raw transcript - Two output modes — Ambient (multi-category SOAP-style synthesis) and Dictation (verbatim with cleanup) - DOCX template engine with inline placeholder tokens, dual-archive export, and anchor-based fill for exact-fidelity Word output ### AI Agent Orchestration LangChain, LangGraph, MCP servers, RAG systems, function calling. I build agents that do real work — not chatbots that summarize, but extraction systems that produce validated, structured data from unstructured sources. ## Anthropic API Cost Optimization I help teams cut Claude/LLM API spend in half on production deployments without sacrificing product quality. I work across the levers that actually move the bill: - **Prompt caching** — designed around real traffic patterns, not just toggled on - **Model tiering** — Sonnet where reasoning matters, Haiku everywhere else - **Context discipline** — lazy-loading tools, skills, and instructions instead of dumping everything upfront - **Tool-call efficiency** — fewer round-trips, smaller payloads I've done this on a medical-document extraction pipeline and a LangGraph trading agent — both running in production with real users and real budgets. If you've inherited an over-budget Claude deployment, I know where to look first. ## Tech I use daily Python, FastAPI, Anthropic/OpenAI/Gemini, Instructor, LangChain, LangGraph, Textract, PyMuPDF, PostgreSQL, Redis, S3, Lambda, SNS/SQS, Docker, Pydantic. 20+ projects delivered on Upwork. I'm strongest when the problem involves turning unstructured documents into clean, structured data at scale.

With nearly 8 years experience working with AI and Machine Learning, I have worked in Vietnam, Singapore and done some freelancing projects for customers from the US, Denmark, and Australia. I have experience working with many state of the art algorithms in NLP. Including prompting and finetuning LLM models; advanced NLP techniques like text data augmentation, distillation, pattern exploiting training to solve text classification and text generation problems. My recent projects are including LLM and RAG Chatbot, Contextual Targeting and Personalized Email Generation, Sentiment Analysis, Summarization, Question Answering. I also take responsibility in designing and deploying the whole AI service architecture on cloud with AWS and GCP services.

Tired of unreliable weather guesses costing you money? I deliver precise, actionable forecasts that help businesses mitigate risks and capitalize on weather patterns. With 5 years in operational meteorology, I bridge the gap between complex atmospheric science and your real-world needs. 1. Why My Forecasts Win? • 90% accuracy rate on high-impact weather events. • Develop optimized weather forecasting systems for HPC/cloud (EC2), built for 24/7 reliability and easy scaling. • Turn NOAA data, WRF models, weather radar and satellite imagery into clear business recommendations. • AI-enhanced modeling that spots risks standard forecasts miss. • Beyond traditional forecasts: using atmospheric science to predict nature's light shows. • Making the unpredictable predictable: advancing clear air turbulence forecasts in aviation. • Assess extreme weather trends with NEX-GDDP across RCP4.5 and RCP8.5. 2. Skills: • Numerical weather prediction models (NWP): WRF, GFS, ECMWF, ICON-D2, AROME,… • Python-driven data analysis • Visualization: weather map, vertical cross section, Skew-T,… by using matplotlib, d3.js,… • Converting and processing meteorological data format file (NetCDF4, GRIB, HDF5,…)

AI Engineer with 5+ years of expertise in computer vision and deep learning. Experienced in leading technical teams, building production ML systems, and implementing MLOps best practices. Passionate about leveraging cutting-edge AI technologies to solve complex business challenges.

I am a data scientist with a strong background in machine learning, artificial intelligence, and data analytics. I has a proven track record of developing and implementing data-driven solutions that have significantly impacted various organizations. My expertise lies in the ability to translate complex data into actionable insights. He is particularly skilled at: + Personalization: real-time personalization solutions for search and recommendation systems. + Machine Learning: adapting machine learning to solve problems, such as customer churn prediction, recommender systems, and promotion optimization. + AI & LLM: applying natural language processing and understanding to build virtual assistants and semantic search engines. + Engineering: deploying machine learning models in production with MLOps practices, data engineering, and data warehouse. With a strong foundation in data science and a passion for innovation, I am eager to tackle new challenges and contribute to the advancement of the field.

If your AI chatbot or RAG system is burning through tokens, slowing down under load, or becoming too expensive to operate, I can help you cut costs and reduce latency while maintaining quality. What I can do for you: - Audit your current OpenAI / LLM usage (identify hidden cost leaks) - Redesign your prompt & context architecture - Build structured memory systems (instead of bloated chat logs) - Optimize RAG pipelines (retrieval, ranking, compression) - Implement model routing to drastically cut costs - Move deterministic logic (filtering, sorting) out of LLM - Set up caching layers (prompt + semantic)