데이터·AI 스킬

Manage the credentials Arize uses to call LLM providers such as OpenAI and Anthropic. Create, inspect, update, or delete connections, configure allowed models and authentication methods, and retrieve connection IDs for use in evaluations.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Connect an LLM provider to Arize
• Review existing connections and allowed models
• Rotate stored provider API keys

Create and analyze AI model evaluation experiments in Arize using the ax CLI, a command-line tool. Collect real model outputs, export results, compare evaluation scores, and identify examples where a new model performs worse.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Comparing models on the same evaluation dataset
• Finding examples that perform worse after a model change
• Exporting experiment results for further analysis

Improve LLM prompts using Arize production traces, evaluation results, and human feedback. Extract current instructions, identify recurring failures, revise the prompt, and compare experiments to measure changes in output quality and catch regressions.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Improve a prompt using failed production traces
• Fix responses that use the wrong output format
• Review unsupported answers in retrieval-based responses

Organize human review in Arize by defining label choices, numeric ranges, or free-text feedback fields. Manage review queues and reviewers, submit annotations for individual records, and apply existing labels to project spans, or individual processing records, in bulk.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Define shared labels and scoring ranges for reviewers
• Create review queues and assign reviewers
• Record feedback on individual queue items

Create and manage Arize datasets for evaluation and experiments using the ax CLI. Import records from files, append examples, inspect dataset versions, and export a chosen version for offline analysis, with checks for field consistency and export completeness.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Creating evaluation datasets from files
• Adding examples to an existing test set
• Inspecting dataset metadata and versions

Design workflows that evaluate and refine AI outputs through self-review, separate generation and evaluation roles, tests, or weighted scoring criteria. Includes patterns for structured feedback, iteration limits, and stopping when quality stops improving.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Adding self-review and revision loops to AI outputs
• Separating generation from evaluation
• Refining generated code using test results

Design an AI agent around its requirements, or audit an existing agent for architectural problems. Deliver justified decisions, verification plans, and a PDF report with a diagram, covering budgets, permissions, stopping rules, and failure handling.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Design an agent from an idea or specification
• Audit an existing agent’s architecture
• Diagnose errors and false completion reports

Add Arize AX tracing to an LLM application, recording how model calls and tool executions connect. Inspect the existing code, select suitable integrations, add instrumentation, and verify that the application runs and traces reach Arize.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Setting up tracing for an LLM application
• Recording tool inputs and outputs in an agent
• Adding Arize export to an existing OpenTelemetry setup

Inspect existing LLM application records in Arize to investigate errors, slow operations, or behavior changes. Export traces, which group related operations, or individual operations and conversation sessions as JSON using the ax command-line tool.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Investigate errors in an LLM application
• Review the inputs and outputs of a conversation
• Find slow operations in recorded traces

Measure denoising latency in SGLang diffusion models and identify slow operations with profiler traces. Compare performance and output quality on the same GPUs, check existing optimizations, and gather evidence before starting low-level kernel work.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Measuring diffusion denoising latency
• Ranking slow operations from profiler traces
• Checking whether existing optimizations address a bottleneck

Choose SGLang Diffusion settings for a specific model, GPU, and memory budget. Compare generation speed, memory use, and output quality to select command-line options, including CPU offloading, multiple GPUs, caching, and reduced precision.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Choosing generation settings for a specific model and GPU
• Reducing GPU memory use to fit a memory budget
• Comparing performance settings across multiple GPUs

Quantize diffusion transformers with NVIDIA ModelOpt, using FP8 or NVFP4 lower-precision numbers, and prepare the saved weights for SGLang Diffusion. Follow a workflow for loading, checking output quality, and comparing speed under matching conditions.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Preparing ModelOpt FP8 weights for SGLang Diffusion
• Checking quality after diffusion model quantization
• Comparing BF16 and quantized generation speed

Analyze LLM serving traces to identify GPU time usage, opportunities to run operations concurrently, and known patterns for combining operations. Produce three consistent tables from saved traces or live captures on supported servers.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Identify GPU operations that dominate inference time
• Compare input processing and token generation
• Trace GPU operations back to Python code

Turn broker trade exports into a report on holding periods, trading frequency, win rate, and realized profit or loss. Diagnose patterns such as holding losers longer, overtrading, chasing rising prices, and fixation on particular price levels, with numerical evidence.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Reviewing win rate, profit or loss, and holding periods
• Checking trading habits with numerical evidence
• Comparing results by month, symbol, or market

Backtest stock and cryptocurrency option strategies using theoretical prices derived from underlying prices. Simulate combinations of option positions, track profit and loss, and export Greeks, measures of sensitivity to price, volatility, and time.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Compare covered call and protective put strategies
• Backtest straddles and iron condors
• Analyze theoretical option prices and sensitivities

Generate trading signals from historical volatility, a measure of past price fluctuations. Rank its current level against recent history, signal long positions at low levels, and exit or short at high levels while retaining positions between thresholds.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Build signals from historical volatility levels
• Adjust calculation windows and percentile thresholds
• Set annualization for cryptocurrency data

Build a pair trading strategy for two related instruments using their price ratio. Measure how far the ratio moves from its recent average, generate opposite buying and selling signals, and close positions when the ratio returns near that average.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Test a pair trade between two stocks in the same industry
• Build a BTC/ETH price ratio strategy
• Set entry and exit thresholds for a pair trade

Compare Chinese A-share sectors using business conditions, price trends, valuation, and fund flows. The skill provides scoring and ranking frameworks to organize sector allocation proposals and trace how demand or cost changes affect connected industries.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Compare business conditions across Chinese A-share sectors
• Rank sectors by price trends, earnings changes, and fund flows
• Trace cost and demand changes through industry supply chains

Analyze crypto perpetual futures funding payments and spot–futures price differences to assess carry trades. Compare exchanges, identify crowded positions, and summarize potential returns alongside risks such as funding reversals and liquidation.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Compare funding rates across crypto exchanges
• Estimate annualized spot–futures price differences
• Assess carry trade returns and risks

Retrieve cryptocurrency prices, candlesticks, order books, funding rates, and open interest from OKX’s public API. Use spot, derivatives, and index data for market research without registering an account or configuring authentication.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Checking cryptocurrency prices and trading volume
• Retrieving historical candlestick data
• Examining funding rates and open interest

Retrieve financial data through Tushare using Python. Find relevant APIs for Chinese stocks, ETFs, funds, bonds, economic indicators, and news, then use documented parameters to obtain pandas DataFrames, tables ready for further analysis.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Retrieving lists of listed Chinese stocks
• Collecting ETF prices and fund size data
• Checking fund holdings and manager information

Apply statistical tests to financial time series, model changing volatility, and diagnose regression problems. Estimate uncertainty in strategy metrics through bootstrap resampling, which repeatedly samples existing data, and organize findings into a statistical testing report.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Check whether price and return series are statistically stable
• Test long-term relationships between two assets
• Diagnose regression problems in factor research

Analyze price and volume data using trend indicators, overbought and oversold measures, and volume changes. Combine these three perspectives into a single long, short, or stand-aside signal, with adjustable calculation periods and thresholds.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Combine several technical indicators into one signal
• Check trends alongside overbought and oversold conditions
• Adjust indicator periods and thresholds

Retrieve price and trading volume data for US, Hong Kong, and Canadian stocks, ETFs, and indices through Yahoo Finance. Also gather company profiles, financial statements, dividends, and ownership data for market research and historical strategy testing.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Collect historical prices across several stock markets
• Review company financial statements and ownership
• Check dividend and stock split histories