I will build rag agents and llm finetuning
AI Engineer Data Scientist
Über diesen Service
Generative AI Developer | AI Engineer | Algorithm Engineer
An AI Engineer from Singapore, leveraging years of hands-on experience working on AI, LLM and Edge-deployed AI application projects to help startups and small businesses integrate AI into their business. Worked professionally in big corps such as Intel, Mediatek, and Jabil - delivering reference AI solutions and designs to tier 1 customers such as Google, Lenovo, and Asus. Has experience in publishing research paper on Financial Reinforcement Learning.
My services / skillsets:
- Local LLM Finetuning - Finetunes LLM up to 14B with Lora using a 5090 GPU.
- LLM RAG - chunking, ingestion, retrieval, multimodal.
- LLM Agents, MCP and workflows.
- LLM Integration and Backend- Openai, Huggingface, Ollama, vLLM
- MLOps/LLMOps - CICD, AWS, MLFlow, Grafana, Langfuse/Langsmith.
- AI-powered app development - React, Tauri.
- AI Algorithms and Optimizations - Transformers, BERT, SPD, MOE, Matformer, quantization.
My tools:
- Frameworks: Unsloth / Langchain / Llamaindex / Huggingface / Openai / Ollama / Vectordb
- Language: Python / C++ / React / Tauri
- IDE: Vscode
Frameworks:
scikit-learn
•
keras
•
PyTorch
•
Panda
•
Andere
Datentyp:
Text
Programmiersprache:
Python
•
SQL
•
Colab
•
MLflow
Tools:
Jupyter-Notizbuch
•
tensorflow
•
Excel
•
MLflow
•
Colab
•
Andere
APIs:
OpenAI
Mein Portfolio
FAQ
Do we need to provide our own documents?
It is optional. A generic file ingestion pipeline will be built if no dataset is provided.
Do you support multimodal RAG?
Image modality can be supported. Two modes of support: Grounded, convert images into text and vectorize as text. Hybrid, vectorize original images in the same vector space as texts.
Do you do Agents or Finetuning LLM?
Depends on the scope and scale of the project. Pricing may differ but it can be discussed.

