zhang long
Community@handsomelong922 · No. 77 Puhe Road, Shenbei New District, Shenyang City, Liaoning Province
I'm a graduate student in Public Health at China Medical University, passionate about enhancing global health. My focus is on epidemiology and disease prevent
Agent Skills by zhang long
Showing 141 vetted skills indexed across 1 GitHub repositories.
handsomelong922Extract text, tables, and data from PDFs using pypdf and pdfplumber.
handsomelong922my-codex-skills-collection
Consolidate Skill Units into a single installable bundle with a root SKILL.md.
handsomelong922dspy
Automate language-model pipeline construction with declarative signatures and modules.
handsomelong922planning-with-files
Create and update Markdown planning files to structure multi-step work.
handsomelong922openrlhf-training
Orchestrate distributed RLHF training for PPO, GRPO, RLOO, and DPO workflows.
handsomelong922using-git-worktrees
Create isolated Git worktrees for feature development with safety checks.
handsomelong922nemo-guardrails
Enforce runtime safety policies for LLM applications with configurable rails.
handsomelong922evaluating-cosmos-policy
Evaluate NVIDIA Cosmos Policy in LIBERO and RoboCasa simulation environments.
handsomelong922llamaindex
Orchestrate data sources and LLM interactions to build RAG pipelines.
handsomelong922deepspeed
Implement and tune DeepSpeed training workflows with ZeRO stages, pipeline parallelism, mixed precision, and MoE on GPU clusters.
handsomelong922csv-data-summarizer
Analyze CSV data to reveal structure, quality, and insights.
handsomelong922crewai-multi-agent
Coordinate multiple autonomous AI agents with memory across sequential or hierarchical workflows.
handsomelong922moe-training
Train Mixture-of-Experts models at scale using DeepSpeed or HuggingFace.
handsomelong922nnsight-remote-interpretability
Run and patch neural network activations with nnsight and NDIF remote execution.
handsomelong922langsmith-observability
Validate SKILL.md frontmatter and generate metadata for Skill units.
handsomelong922audiocraft-audio-generation
Generate music and sound effects from textual prompts using PyTorch.
handsomelong922ai-search-hub
Coordinates AI search queries across platforms via Chrome DevTools sessions.
handsomelong922nemo-curator
GPU-accelerates multimodal dataset curation with quality filtering and deduplication.
handsomelong922huggingface-accelerate
Unify distributed training across single and multi-GPU environments with accelerate.
handsomelong922self-improvement
Log learnings and corrections to .learnings markdown files for future sessions.
handsomelong922test-driven-development
Write failing tests before production code using the Red-Green-Refactor cycle.
handsomelong922chroma
Store and retrieve embeddings with metadata using a vector database.
handsomelong922prompt-guard
Detect jailbreak and prompt-injection attempts in LLM applications.
handsomelong922evaluating-llms-harness
Benchmarks LLMs across 60+ academic tasks producing standardized performance metrics.