3. Data, Science & AI

Automates schema mapping and synchronization for embedded documents, ensuring data consistency across structures.

Validates data science analysis results, enforces reproducibility checks, and requires user acceptance before finalizing outputs.

Provides expert guidance on M3's gokart, a Python ML pipeline tool, covering type-safe pipeline design, testing, and Pandera integration for robust machine learning workflows.

Queries a worklog database to retrieve context, knowledge, errors, and historical data for developer reference.

Guides creation, configuration, and use of Opencode sub-agents and the `delegate_to_agent` tool for agent delegation in AI systems.

Analyzes Claude Code usage patterns, detects repetitive tasks, and visualizes productivity trends by time of day and frequency.

Creates optimized prompts for Agent-to-Agent pipelines with research, planning, and execution stages to enable seamless multi-stage AI workflows.

Extracts Schedule of Investments data including debt and equity holdings from BDC SEC filings via XBRL taxonomy.

Monitors AI model context window usage and offers optimization recommendations to improve response speed and quality during sessions.

Standardizes ECharts-based visualizations and themes for InsightPulseAI dashboards, Superset plugins, and AntD/M3 UI frameworks.

Generates weekly reports analyzing Claude Code usage patterns, including project time distribution, work types, and Jira issue progress from session logs.

Converts drug, PICO, and CQ input modes into unified structured data for RWD research planning.

Provides efficient Solana blockchain data access through Alchemy API, including NFTs, transactions, and asset queries with error handling and caching.

Designs specification for extracting clinical events from narrative medical texts using NLP and regex patterns, including quality management.

Enables effective prompt design for Claude API, including system prompts, context management, and instruction optimization to enhance AI interactions.

Creates, validates, and debugs YAML configurations for axolotl-rs model fine-tuning workflows.

High-performance processing of genomic data (VCF, FASTA, BED) with streaming, interval joins, and variant annotation features.

Copies text extracted from image files to clipboard for transcription and OCR tasks.

Creates computer-executable specifications for phenotype definitions and lab data utilization in research studies, including validation plans.

Enables the agent to save, recall, and organize user-provided information within its memory system.

Automates syncing of Hevy workout data and delivers AI-driven fitness insights for exercise, volume, and personal records.

Performs data analysis and visualization using Python pandas and visualization libraries to extract actionable insights.

Adds new exercises or equipment to system catalogs with automatic duplicate checks, normalization, alias creation, and referential integrity.

Executes Python code in a secure isolated environment for data calculations, file generation, and dataset processing.