3. Data, Science & AI

Refines prompts for Anthropic's Claude models (Opus, Sonnet, Haiku) using best practices to improve task execution.

Provides advanced DSPy modules for building AI agent systems with ReAct agents, tool integration, and multi-step reasoning capabilities.

Enables communication with Jules AI, an autonomous coding agent, for automated code generation and development assistance.

Optimizes DSPy programs in skills-fleet through a structured 3-phase workflow including training, monitoring, and versioning for improved AI application performance.

Configures DSPy framework for language model setup, caching, and version management to optimize AI application development.

Provides foundational DSPy tools for designing signatures, using modules like Predict and ChainOfThought, and building simple AI programs.

Provides advanced techniques for optimizing LLM prompts to enhance performance, reliability, and controllability in production environments.

Extracts text content from PDF documents for data processing and analysis using PyPDF2 or pdfplumber libraries.

Executes ComfyUI workflows to generate images, video, and audio content using AI models.

Expert-level patterns for building production chatbots with Vercel AI SDK v5, including persistence, generative UI, and Anthropic provider integration.

Enables seamless LLM API integration with prompt templates, error handling, and provider abstraction for AI features.

Enriches GitHub database entries in Notion with metadata from GitHub API, README, YouTube videos, and linked articles, establishing bidirectional relationships in YouTube and Web Resources databases with freshness tracking.

Enables comprehensive spreadsheet creation, editing, data analysis, and visualization for .xlsx, .xlsm, .csv, and .tsv files.

Enables discovery, retrieval, and enhancement of AI prompts and templates for optimized interaction with generative models.

Displays statistical correlation between predicted and actual skill usage to assess routing effectiveness.

End-to-end machine learning model development including dataset preparation, SFT, DPO, LoRA fine-tuning, evaluation, and deployment optimization.

Collects, analyzes, and prioritizes SEMO user feedback issues using Supabase database for product management.

High-performance CLI tool for splitting, merging, and analyzing SQL dump files across multiple database systems, with features for data extraction, schema visualization, and safe migration.

Monitors real-time token usage in AI conversations, predicts when token limits will be reached, and provides proactive warnings to prevent service interruptions.

Designs and operates analytical data platforms with ETL/ELT pipelines, dimensional modeling, and data quality management.

Enables product managers to define metrics, build analytics funnels/cohorts, and translate data insights into product actions through SQL-style analysis plans and dashboards.

Provides AI-powered knowledge base operations including querying, daily report generation, agent checks, and index rebuilding for efficient information retrieval.

Creates a visual year-end summary of repository activity, contributions, and statistics for code repositories.

Extracts structured data (emails, URLs, dates, entities) from unstructured text using schema validation and anti-hallucination safeguards.