职业

数据科学家

24533 个技能

AI 推荐

24533 个技能

Rag Engineer

Expert in building Retrieval-Augmented Generation systems. Masters embedding models, vector databases, chunking strategies, and retrieval optimization for LLM applications. Use when: building RAG, vector search, embeddings, semantic search, document retrieval.

freshwaterbruce20

实验室工具

Ai Ml Experiment Rigor

Design and audit AI/ML experiments for comparability, reproducibility, and realistic compute use. Use when the user needs help defining baselines, ablations, metrics, data splits, seed handling, experiment tracking, statistical interpretation, compute budgets, or a reproducible experiment spec before running large jobs.

frinfo7020

机器学习

Model Fine Tuning

Guides when and how to fine-tune AI models for specific tasks, domains, and behaviors. Covers data preparation, training strategies, evaluation metrics, and deployment across OpenAI, Google Vertex AI, and open-source (LoRA/PEFT). Use when deciding whether to fine-tune, preparing training data, selecting hyperparameters, evaluating model quality, or choosing between fine-tuning and prompt engineering.

frvnkfrmchicago0

调试

Mlstack

ML and data analysis team as slash commands. Ten specialist modes: PI analysis review, biostatistician methods review, ML architecture review, notebook review, exploratory data analysis, feature engineering, adversarial model critique, performance optimization, analytics retrospective, and analysis shipping. Each mode activates a different expert persona with domain-specific checklists, critique patterns, and quality standards.

tim-krausz0

机器学习

ML Architect Plan Review

ML Architect-mode plan review. Challenge the modeling strategy across data modalities: raw signals (pixels, waveforms, sequences), learned representations (embeddings, latent spaces), and structured features (tabular, graph). Reviews architecture choices, representation strategy, training dynamics, compute-performance tradeoffs, and deployment feasibility. The person who asks "why are you hand-engineering features when you could learn them end-to-end?" AND "why are you training a 200M parameter model when a gradient-boosted tree on 12 features gets you 95% of the way there?"

tim-krausz0

机器学习

Grpo Rl Training

Expert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training

future13140

数据科学家

Rag Engineer

Ai Ml Experiment Rigor

Model Fine Tuning

Mlstack

ML Architect Plan Review

Grpo Rl Training

数据科学家

Rag Engineer

Ai Ml Experiment Rigor

Model Fine Tuning

Mlstack

ML Architect Plan Review

Grpo Rl Training

Finding Last Valid Versioned Snapshot

Recovering Central Document From Embeddings

Wrds Data

Xlsx Analyser

Skill: table-text numerical reasoning qa

Compbio Workflow

Rag Learning Optimizer

Analyze

Kv Lab Research

Source-Sink Security Analysis

Viz Research

Csv Pipeline

Data Pipeline Reviewer

Pytorch Build Resolver

Senior ML Engineer

Senior Computer Vision Engineer

Strategist Predictions

Analyst Revenue Analytics

Autoresearch

OpenEnv Environments

Seaborn

Embedding Strategies

Ai Engineering Toolkit

Hugging Face Model Trainer