Data & AI skills

Convert human-written CLAUDE.md instructions into a structured label format for AI agents. Use it to review lengthy rules, reduce instruction overhead, and prepare prompts for migration between AI tools while preserving the intended rules.

• Restructuring a lengthy CLAUDE.md
• Reviewing instruction token usage
• Preparing rules for migration between AI tools

Design AI agents that use tools, plan tasks, retain relevant context, and recover from failures. Choose execution patterns, permission boundaries, memory handling, and runtime limits, and assess whether coordinating multiple agents is justified for the task.

• Choosing an execution loop for a tool-using agent
• Designing memory and recovery for long-running tasks
• Setting tool permissions, runtime limits, and cost caps

Create consistent records for evaluating AI agents, including task success, invalid tool calls, token use, and actual costs. Produce CSV tables, database definitions, and JSON validation schemas to support reproducible comparisons and checks for performance declines.

• Recording agent evaluation results in a shared format
• Comparing a candidate agent with a baseline
• Tracking actual cost per solved task

Plan AI application work across six areas: prompt evaluation, context allocation, retrieval-based answer generation, agent security, evaluation pipelines, and product strategy. This overview explains the approaches; the individual workflow packages, test cases, and CI templates are not included.

• Planning prompt evaluation before deployment
• Allocating context capacity and designing retrieval-based answers
• Planning an authorized AI agent security assessment

Helps design focused AI products built on APIs such as OpenAI and Anthropic. Covers product strategy, prompts, model selection, user experience, usage tracking, and cost management for tools aimed at solving specific customer problems.

• Planning an AI SaaS product for a specific problem
• Designing prompts for product features
• Balancing model choices, costs, and user experience

Store project knowledge and design decisions in persistent memory that AI agents can search and read. New findings sync automatically to Markdown documentation, while usage statistics help you inspect the memory bank.

• Save architecture decisions for later tasks
• Find previously recorded implementation patterns
• Keep agent memory synced with project documentation

Improve an existing AI agent by analyzing execution records, user feedback, and performance metrics. Refine prompts and workflows, compare original and revised versions through tests, and plan staged releases with monitoring and rollback when results worsen.

• Investigating recurring agent failures and user corrections
• Refining prompts, examples, and output formats
• Comparing original and revised agents with A/B tests

Plan an A/B test to compare product changes, with a clear hypothesis, variants, sample estimates, and tracking checks. Define success metrics and stopping rules before launch, then document results and decisions.

• Planning a signup form A/B test
• Estimating sample size and test duration
• Checking event tracking before launch

Create a dedicated AI subagent for recurring work, with a defined role, working process, output format, and constraints. Package its files in a plugin and optionally draft a companion skill that routes suitable tasks to it.

• Creating a dedicated code review agent
• Turning repeated instructions into an agent persona
• Adding related agents to an existing plugin

Design criteria for using an LLM to judge AI responses through scoring or comparisons between two answers. Includes checks for order and length bias, plus validation against human judgments to assess evaluation reliability.

• Build an automated evaluation pipeline for AI responses
• Compare responses after changing a prompt or model
• Define shared scoring criteria for reviewers

Route a goal through acc's scored-memory workflow, which uses evaluated stored information. Handle returned reasoning frames, submit proposals with confidence estimates, and present final answers with reference IDs before recording the eventual outcome.

• Submit a goal to acc's scored-memory workflow
• Review a returned frame and submit a proposal
• Present answers with commitment and reference IDs

Provides persistent project memory for AI agents through a local MCP server, a tool interface for agents. Stores and searches architecture decisions and implementation patterns, syncs with project documentation, and includes a dashboard for viewing memory usage.

• Save architecture decisions for later reference
• Search previously recorded implementation patterns
• Retrieve design notes by key

Design agent memory that preserves facts, preferences, and conversation history across sessions. Plan storage, retrieval, and retention separately for each organization or user, with checks for unauthorized access, outdated information, conflicting records, and retrieval quality.

• Designing memory that remembers preferences across sessions
• Separating conversation history by organization and user
• Investigating retrieval of outdated or conflicting records

Build persistent agents on Azure AI Foundry with the Microsoft Agent Framework Python SDK. Connect functions, search, and MCP tools, preserve conversation context, stream responses, and return data in a defined structure.

• Building an agent that preserves conversation context
• Connecting Python functions and hosted search tools
• Displaying responses as they arrive

Write and review SQL for reports and metrics in a data warehouse, a store of analytical data. Applies rules for explicit columns, date filters, aggregate names, and limited exploratory results, returning SQL with a brief explanation.

• Write SQL for report and dashboard metrics
• Calculate counts, sums, and rolling metrics
• Combine data from multiple tables

Find ranked crypto tokens and trader addresses by social buzz, search popularity, capital inflows, or trading profit and loss. Compare results across supported blockchains, with time windows and filters available for relevant rankings.

• Find tokens attracting social buzz
• Compare tokens by net capital inflows
• Filter trending tokens by market size

Explore World Cup match predictions, related news, and market probabilities. Adjust model factors to compare possible outcomes, then optionally hand off to Binance Agentic Wallet for prediction-market trading, where orders require an explicit choice and confirmation.

• Review AI predictions for a World Cup match
• Compare model probabilities with market probabilities
• Recalculate probabilities with adjusted factors

Configure supervised fine-tuning for models that process images and text. Define which components remain unchanged, set LoRA adapters for updating selected model weights, choose image resolution limits, and investigate training runs whose evaluation results fail to improve.

• Configure training for an image-and-text task
• Decide whether to train vision layers
• Investigate training with flat evaluation scores

Prepare GRPO reinforcement learning configurations for tasks with programmatically checkable answers, such as math, code, and structured output. Design and inspect reward functions, then select training variants when runs show instability or exploit flaws in scoring.

• Preparing GRPO training for tasks with automatic correctness checks
• Designing rewards for output format and correctness
• Inspecting reward scores before training

Turn graded execution traces and production logs into model training examples and preferred-versus-rejected response pairs. Select useful records, remove sensitive information, exclude evaluation examples, and preserve links to the original runs.

• Convert high-scoring successful runs into training examples
• Use expert-corrected failures as training examples
• Pair passing and failing responses to the same task

Prepare an evaluation framework before fine-tuning a language model. Build labeled examples, checks for each failure type, and a base-model benchmark, while calibrating model-based judges against human labels to support comparisons after training.

• Build an evaluation set from agent traces
• Create separate checks for each failure type
• Calibrate an LLM judge against human labels

Design recommendation and ranking systems that select the best items for a user or situation. Organize retrieval, enrichment, filtering, scoring, selection, and background actions into six stages, then generate a runnable scaffold for your chosen stack.

• Designing a personalized content feed
• Reranking retrieved passages for RAG
• Prioritizing suggested tasks

Design machine learning workflows from data preparation through training, validation, deployment, and monitoring. Use templates and guidance to connect stages, track versions and experiments, handle failures, and plan gradual releases with rollback options.

• Designing an end-to-end machine learning pipeline
• Building reproducible training workflows
• Comparing models before production deployment

Evaluate whether a fine-tuned model is ready for release by checking training data, general capability loss, comparisons with the base model, and staged production rollout. Produces an evidence-based promotion or rejection report with one prioritized remedy when rejected.

• Deciding whether to release a fine-tuned model
• Checking whether training reduced general capabilities
• Reassessing a model against updated reference evaluations