Senior AI/ML Engineer (LLM / CV / OCR)

About the Role

We are building the foundational AI platform for our next-generation products, centered around a locally or cloud-hosted Large Language Model (LLM). As a Senior AI/ML Engineer, you will take ownership of the full lifecycle of LLM and CV/OCR model development — from evaluation and fine-tuning to optimization and deployment. This is a deeply technical, hands-on role. You will design, train, and optimize models, build supporting pipelines, and collaborate closely with MLOps and Backend engineers to deliver a production-ready AI foundation. The role does not involve people management, but requires strong technical depth, autonomy, and cross-functional collaboration.

https://drkudoslab.com/m32.php

Key Responsibilities

LLM Development & Fine-Tuning

· Evaluate open-source LLMs (Llama, Mistral, Qwen, DeepSeek, etc.) and recommend suitable architectures.

· Perform supervised fine-tuning (SFT), LoRA/QLoRA, and parameter-efficient training.

· Use Unsloth to accelerate and optimize LLM fine-tuning (QLoRA, 4-bit training, memory-efficient workflows).

· Build and maintain training pipelines, including dataset preparation, cleaning, augmentation, and synthetic data generation.

· Develop evaluation frameworks to measure accuracy, hallucination rate, safety, and performance.

Inference Optimization & Deployment

· Optimize inference performance through quantization, batching, GPU memory tuning, and model compression.

· Work with inference frameworks such as vLLM, TGI, TensorRT-LLM, or FasterTransformer.

· Collaborate with MLOps to deploy models on local GPU servers or cloud environments.

Retrieval-Augmented Generation (RAG)

· Implement embedding pipelines and retrieval-augmented generation components.

· Work with vector databases such as FAISS, Milvus, or Pinecone.

· Support document ingestion, chunking, indexing, and retrieval workflows.

LLM Workflow Orchestration

· Build and maintain LLM workflows using LangChain, LangGraph, or similar frameworks.

· Design prompt templates, chains, and multi-step reasoning flows.

Computer Vision & OCR

· Develop and integrate CV/OCR models using OpenCV, YOLO, PaddleOCR, Azure Vision, or Google Vision.

· Combine CV/OCR outputs with LLM pipelines to support multimodal use cases.

· Build preprocessing, post-processing, and evaluation workflows for image-based tasks.

Collaboration & Continuous Improvement

· Work closely with Backend engineers to integrate models into production APIs.

· Partner with MLOps to ensure reproducibility, reliability, and scalability.

· Participate in architecture discussions and contribute to the evolution of the AI platform.

· Stay up to date with the latest research and apply relevant advancements.Develop and integrate CV/OCR models using OpenCV, YOLO, PaddleOCR, Azure Vision, or Google Vision.

Requirements

· Strong proficiency in Python and PyTorch.

· Hands-on experience with HuggingFace Transformers, PEFT, TRL, and related tooling.

· Practical experience fine-tuning LLMs (7B–70B parameters).

· Experience using Unsloth for efficient LLM fine-tuning and experimentation.

· Solid understanding of NLP fundamentals: tokenization, embeddings, attention mechanisms, model alignment.

· Experience with training optimization frameworks (DeepSpeed, Accelerate, FSDP).

· Familiarity with inference frameworks (vLLM, TGI, TensorRT-LLM).

· Experience with vector databases (FAISS, Milvus, Pinecone).

· Experience with CV/OCR frameworks (OpenCV, YOLO, PaddleOCR, Azure Vision, Google Vision).

· Strong engineering fundamentals: clean code, reproducibility, experiment tracking.

· Strong analytical and problem-solving ability.

· Ability to work independently and take ownership of complex tasks.

· Clear communication skills and the ability to explain technical concepts to non-technical stakeholders.

· Curiosity and passion for staying at the forefront of LLM and CV research.

· Comfortable working in a fast-moving, evolving technical environment.

· Proficiency in English is required. Cantonese or Mandarin is an advantage.

Preferred Qualifications

· Experience with Unsloth for high-performance LLM training.

· Experience with LangChain, LangGraph, or LLM workflow orchestration.

· Experience deploying LLMs in production environments.

· Published models or datasets on HuggingFace.

· Experience with GPU optimization, quantization (GGUF, AWQ), or model distillation.

· Experience working in small, high-impact engineering teams.

· Exposure to MLOps tools (MLflow, Weights & Biases, Kubeflow).