Phạm Phú Ngọc Trai
Community@jayll1303 · HCM Viet Nam
FPTU K16 @JayLL-vn
Agent Skills by Phạm Phú Ngọc Trai
Showing 31 vetted skills indexed across 1 GitHub repositories.
jayll1303semantic-router
Routes user queries to handlers using semantic vector matching and configurable encoders/indexes.
jayll1303notebook-workflows
Automate Jupyter and Colab notebook creation and editing via nbformat and nbconvert.
jayll1303ml-brainstorm
Brainstorm ML/AI decisions and recommend paths from repository context.
jayll1303openai-audio-api
Build OpenAI-compatible TTS endpoints with streaming and batch inference.
jayll1303hf-transformers-trainer
Fine-tune and align local LLMs with HuggingFace Trainer, PEFT, and TRL.
jayll1303sherpa-onnx
Perform offline speech processing with ONNX models via sherpa-onnx.
jayll1303opentelemetry
Instrument Python and Node.js apps with OpenTelemetry to collect traces, metrics, and logs.
jayll1303hf-speech-to-speech-pipeline
Orchestrate real-time speech-to-speech pipelines connecting VAD, STT, LLM, and TTS components.
jayll1303python-project-setup
Bootstrap Python projects with uv and a single pyproject.toml.
jayll1303paddleocr
Train, fine-tune, and export PaddleOCR models for text detection and recognition.
jayll1303ollama-local-llm
Manage local LLMs via Ollama CLI and API with Modelfile creation.
jayll1303python-quality-testing
Add type annotations, contract docstrings, Hypothesis tests, and mutmut mutation testing to Python ML code.
jayll1303hf-hub-datasets
Download and upload HuggingFace Hub models and datasets with authentication.
jayll1303tensorrt-llm
Convert HuggingFace checkpoints into optimized TensorRT-LLM engines with quantization.
jayll1303text-embeddings-rag
Generate local embeddings and build retrieval-augmented generation pipelines with sentence-transformers and FAISS.
jayll1303arxiv-reader
Fetch and analyze arXiv papers via HTML-rendered pages.
jayll1303aie-skills-installer
Detect technology signals in Python and ML repositories to recommend and install matching AIE-Skills.
jayll1303experiment-tracking
Deploy self-hosted MLflow and Weights & Biases trackers for ML experiments.
jayll1303llama-cpp-inference
Run local GGUF model inference and OpenAI-compatible serving with llama.cpp.
jayll1303sglang-serving
Launch and tune SGLang servers with constrained decoding and RadixAttention.
jayll1303vllm-tgi-inference
Deploy local vLLM or HuggingFace TGI inference servers with OpenAI-compatible APIs.
jayll1303docker-gpu-setup
Build GPU-enabled Docker containers with NGC base images and uv.
jayll1303ultralytics-yolo
Train and export Ultralytics YOLO models for detection and tracking.
jayll1303k2-training-pipeline
Train ASR and TTS models with k2, icefall, and lhotse pipelines.