ihatesea69

ihatesea69/kiro-kit

Command

⚡⚡⚡ Intelligent plan creation with prompt enhancement.

ihatesea69/kiro-kit

Command

Prepare and execute a release with version bump and changelog.

ihatesea69/kiro-kit

Command

Scout the codebase for files relevant to a task.

ihatesea69/kiro-kit

Command

Run the test suite with coverage reporting.

ihatesea69/kiro-kit

Command

Utilize tools of Model Context Protocol (MCP) servers.

ihatesea69/kiro-kit

Command

⚡ Review recent changes and wrap up the work.

ihatesea69/kiro-kit

Skill Claude CodeCodex

Create aesthetically beautiful interfaces following proven design principles. Use when building UI that needs visual polish, micro-interactions, or design system refinement.

ihatesea69/kiro-kit

Skill Claude CodeCodex

Process and generate multimedia content using AI APIs. Use when working with images, audio, video, or document analysis.

ihatesea69/kiro-kit

Skill Claude CodeCodex

Deploy mobile apps to App Store and Google Play. Use when configuring builds, signing, CI/CD pipelines, or preparing store submissions.

ihatesea69/kiro-kit

Skill Claude CodeCodex

Build robust backend systems with Node.js, Python, or Go. Use when implementing APIs, services, middleware, or server-side logic following production-grade patterns.

ihatesea69/kiro-kit

Skill Claude CodeCodex

Implement authentication and authorization with modern patterns including OAuth 2.1, JWT, sessions, RBAC, and multi-factor authentication. Use when building auth systems.

ihatesea69/kiro-kit

Skill Claude CodeCodex

Browser automation and debugging using Puppeteer for API testing, screenshot capture, and web scraping. Use when you need to interact with web interfaces programmatically.

ihatesea69/office-powerpoint-mcp-server

MCP server Claude CodeCodexCursor

MCP Server for PowerPoint manipulation using python-pptx - Consolidated Edition. Runs locally from the office-powerpoint-mcp-server Python package.

ihatesea69/HieuNghi-AI-Skills

Instructions file CodexOpenCode

Instructions for ihatesea69/HieuNghi-AI-Skills, covering hieunghi-ai-kits — central agent instruction, what is this repository?, skill collections, 📝 devtoskills/ — dev.to blog writing kit and ⚛️ engineeringskills/ — engineering best practices kit.

ihatesea69/HieuNghi-AI-Skills

Skill Claude CodeCodex

Implements and trains LLMs using Lightning AI's LitGPT with 20+ pretrained architectures (Llama, Gemma, Phi, Qwen, Mistral). Use when need clean model implementations, educational understanding of architectures, or production fine-tuning with LoRA/QLoRA. Single-file implementations, no abstraction layers.

ihatesea69/HieuNghi-AI-Skills

Skill Claude CodeCodex

State-space model with O(n) complexity vs Transformers' O(n²). 5× faster inference, million-token sequences, no KV cache. Selective SSM with hardware-aware design. Mamba-1 (dstate=16) and Mamba-2 (dstate=128, multi-head). Models 130M-2.8B on HuggingFace.

ihatesea69/HieuNghi-AI-Skills

Skill Claude CodeCodex

Educational GPT implementation in 300 lines. Reproduces GPT-2 (124M) on OpenWebText. Clean, hackable code for learning transformers. By Andrej Karpathy. Perfect for understanding GPT architecture from scratch. Train on Shakespeare (CPU) or OpenWebText (multi-GPU).

ihatesea69/HieuNghi-AI-Skills

Skill Claude CodeCodex

RNN+Transformer hybrid with O(n) inference. Linear time, infinite context, no KV cache. Train like GPT (parallel), infer like RNN (sequential). Linux Foundation AI project. Production at Windows, Office, NeMo. RWKV-7 (March 2025). Models up to 14B parameters.

ihatesea69/HieuNghi-AI-Skills

Skill Claude CodeCodex

Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs with Float8, torch.compile, and distributed checkpointing.

ihatesea69/HieuNghi-AI-Skills

Skill Claude CodeCodex

Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds. Supports BPE, WordPiece, and Unigram algorithms. Train custom vocabularies, track alignments, handle padding/truncation. Integrates seamlessly with transformers. Use when you need high-performance…

ihatesea69/HieuNghi-AI-Skills

Skill Claude CodeCodex

Language-independent tokenizer treating text as raw Unicode. Supports BPE and Unigram algorithms. Fast (50k sentences/sec), lightweight (6MB memory), deterministic vocabulary. Used by T5, ALBERT, XLNet, mBART. Train on raw text without pre-tokenization. Use when you need multilingual support, CJK languages, or…

ihatesea69/HieuNghi-AI-Skills

Skill Claude CodeCodex

Expert guidance for fine-tuning LLMs with Axolotl - YAML configs, 100+ models, LoRA/QLoRA, DPO/KTO/ORPO/GRPO, multimodal support.

ihatesea69/HieuNghi-AI-Skills

Skill Claude CodeCodex

Expert guidance for fine-tuning LLMs with LLaMA-Factory - WebUI no-code, 100+ models, 2/3/4/5/6/8-bit QLoRA, multimodal support.

ihatesea69/HieuNghi-AI-Skills

Skill Claude CodeCodex

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.