ReddRadar
Topic: artificial-intelligence
56 skills in this topic.
-
evaluating-llms-harness
Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.
OpenRaiser/NanoResearch 689
-
unsloth
Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization
OpenRaiser/NanoResearch 689
-
huggingface-accelerate
Simplest distributed training API. 4 lines to add distributed support to any PyTorch script. Unified API for DeepSpeed/FSDP/Megatron/DDP. Automatic device placement, mixed precision (FP16/BF16/FP8). Interactive config, single launch command. HuggingFace ecosystem standard.
OpenRaiser/NanoResearch 689
-
creative-thinking-for-research
Applies cognitive science frameworks for creative thinking to CS and AI research ideation. Use when seeking genuinely novel research directions by leveraging combinatorial creativity, analogical reasoning, constraint manipulation, and other empirically grounded creative strategies.
OpenRaiser/NanoResearch 689
-
ml-paper-writing
Write publication-ready ML/AI/Systems papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM, OSDI, NSDI, ASPLOS, SOSP. Use when drafting papers from research repos, structuring arguments, verifying citations, or preparing camera-ready submissions. Includes LaTeX templates, reviewer guidelines, and citation verification workflows.
OpenRaiser/NanoResearch 689
-
nanoresearch-ideation
Search academic literature and generate research hypotheses
OpenRaiser/NanoResearch 689
-
nanoresearch-planning
Produce an experiment blueprint from a research hypothesis
OpenRaiser/NanoResearch 689
-
nanoresearch-writing
Draft a LaTeX research paper from all previous stage outputs
OpenRaiser/NanoResearch 689
-
autoresearch
Orchestrates end-to-end autonomous AI research projects using a two-loop architecture. The inner loop runs rapid experiment iterations with clear optimization targets. The outer loop synthesizes results, identifies patterns, and steers research direction. Routes to domain-specific skills for execution, supports continuous agent operation via Claude Code /loop and OpenClaw heartbeat, and produces research presentations and papers. Use when starting a research project, running autonomous experiments, or managing a multi-hypothesis research effort.
OpenRaiser/NanoResearch 689
-
academic-plotting
Generates publication-quality figures for ML papers from research context. Given a paper section or description, extracts system components and relationships to generate architecture diagrams via Gemini. Given experiment results or data, auto-selects chart type and generates data-driven figures via matplotlib/seaborn. Use when creating any figure for a conference paper.
OpenRaiser/NanoResearch 689
-
nanoresearch-experiment
Generate a Python code skeleton from an experiment blueprint
OpenRaiser/NanoResearch 689
-
brainstorming-research-ideas
Guides researchers through structured ideation frameworks to discover high-impact research directions. Use when exploring new problem spaces, pivoting between projects, or seeking novel angles on existing work.
OpenRaiser/NanoResearch 689
-
ray-data
Scalable data processing for ML workloads. Streaming execution across CPU/GPU, supports Parquet/CSV/JSON/images. Integrates with Ray Train, PyTorch, TensorFlow. Scales from single machine to 100s of nodes. Use for batch inference, data preprocessing, multi-modal data loading, or distributed ETL pipelines.
OpenRaiser/NanoResearch 689
-
peft-fine-tuning
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.
OpenRaiser/NanoResearch 689
-
ml-training-recipes
Battle-tested PyTorch training recipes for all domains — LLMs, vision, diffusion, medical imaging, protein/drug discovery, spatial omics, genomics. Covers training loops, optimizer selection (AdamW, Muon), LR scheduling, mixed precision, debugging, and systematic experimentation. Use when training or fine-tuning neural networks, debugging loss spikes or OOM, choosing architectures, or optimizing GPU throughput.
OpenRaiser/NanoResearch 689
-
skypilot-multi-cloud-orchestration
Multi-cloud orchestration for ML workloads with automatic cost optimization. Use when you need to run training or batch jobs across multiple clouds, leverage spot instances with auto-recovery, or optimize GPU costs across providers.
OpenRaiser/NanoResearch 689
-
implementation-strategy
Analyze change impact across services before starting implementation. Trigger before modifying runtime, API, or cross-service code — maps changed files to service ownership, outputs the dependency graph between affected services, and lists upstream/downstream sync points that may need coordinated updates.
tgoai/tgo 420
-
functional-verification
Use tgo-cli (staff) and tgo-widget-cli (visitor) to verify API and service changes at runtime, beyond static lint/build checks. Trigger after modifying backend API endpoints, service logic, chat flow, agent config, knowledge/RAG, workflow, or platform integration — requires local services to be running. Auto-detects changed services from git diff and runs the corresponding CLI smoke tests (system info, CRUD listing, chat e2e).
tgoai/tgo 420
-
cross-service-sync
Detect schema, type, or API interface changes and find files in other services that may need synchronized updates. Trigger when modifying files under schemas/, types/, or interfaces/ directories — extracts changed class/type names from the diff and searches all other services for references, outputting a list of potentially stale consumers.
tgoai/tgo 420
-
local-services
Start, stop, and check local development services with intelligent minimum-set management. Trigger when services need to be running for functional verification or manual testing — starts the minimum required set (infrastructure + tgo-api + tgo-ai) by default, with optional extras auto-detected from git diff or specified manually. Includes status checking and graceful shutdown.
tgoai/tgo 420
-
pr-draft-summary
Generate a PR change summary grouped by service after work is finished. Trigger when wrapping up a task and ready to commit or create a pull request — reads git diff and git log, groups changed files by service directory, and outputs a markdown-formatted summary with file lists, commit history, and diff stats.
tgoai/tgo 420
-
streaming-protocol-check
Check streaming protocol consistency across all producer and consumer services. Trigger when modifying code related to streaming, SSE, WuKongIM, json-render, or MixedStreamParser — lists all files in tgo-ai (producer), tgo-api (relay), tgo-web, tgo-widget-js, and tgo-widget-miniprogram (consumers) that handle the same protocol and may need coordinated updates.
tgoai/tgo 420
-
code-change-verification
Run lint, type-check, and build verification for changed services after code modifications are complete. Trigger after any code change to repos/*/ — auto-detects which services were touched via git diff and runs the appropriate static checks (mypy/flake8 for Python, type-check/lint/build for TypeScript, go vet for Go).
tgoai/tgo 420
-
db-migration-check
Verify that SQLAlchemy model changes have corresponding Alembic migration files. Trigger when any file under models/*.py or models/**/*.py is modified — checks git diff for model changes and confirms matching alembic/versions/*.py files exist, failing if a service has model changes without a migration.
tgoai/tgo 420