3. Data, Science & AI

Guides users on StickerNest's self-improving AI system, covering prompt versioning, reflection loops, and AI evaluation for continuous model optimization.

Assists in developing, training, evaluating, and deploying machine learning models using MLOps practices.

Manages AI reflection cycles for evaluating model quality, improving prompts, and analyzing performance in StickerNest.

Provides persistent semantic memory for AI agents, enabling context ingestion, search, and consolidation across sessions via Qdrant and Ollama infrastructure.

Manages user preferences and adapts system behavior through learning from historical feedback, enabling a 'tell me once' experience.

Persists decisions and learnings to the Mother-Harness long-term memory system for AI knowledge retention and retrieval.

Automates correction of common data errors in test files: type coercion, range format conversion, and schema migration for version upgrades.

Performs data science-focused code reviews with quality checks, data leakage detection, and statistical validity verification using ruff, pyright, and pytest.

Manages Supabase PostgreSQL databases including schema design, migrations, RLS policies, and performance optimization for enterprise applications.

Retrieves past decisions stored in the Mother-Harness long-term memory system for analysis or reference.

Enables step-based orchestration of machine learning pipelines for training and prediction workflows.

Expertly crafts and optimizes prompts for LLMs and AI agents to minimize hallucinations and maximize response reliability.

Integrates Langfuse observability with AWS Strands Agents for tracing, monitoring, and debugging AI agent applications in production.

Analyzes datasets to identify trends and delivers actionable recommendations for data visualization implementation.

Generates safe SELECT queries for PostgreSQL from natural language input with automatic schema detection across database sizes.

Generates semantic layer documentation for data models, business metrics, and data dictionaries to enable AI-assisted analytics and clear data context.

Framework for optimally selecting specialized AI agents based on task characteristics using decision tree logic to ensure the right agent is chosen for delegation.

Provides programmatic access to TheSportsDB API for retrieving sports data including teams, players, leagues, and live scores.

Enables creation, editing, and analysis of spreadsheets with formula support, formatting, and data visualization for data processing tasks.

Integrates OpenStreetMap street networks into Vehicle Routing Problems (VRP) for real-world logistics scenarios, enabling network-based distance computation and instance creation.

Builds autonomous AI agents using AWS's Strands Agents SDK, supporting multi-agent orchestration, custom tools, and integration with model providers like Bedrock and OpenAI.

Provides autonomous management of Readwise highlights, reading data search, article saving, and pattern analysis via Readwise APIs.

Reviews SQL queries for correctness, performance, and adherence to best practices, identifying logical errors and anti-patterns.

Collects and aggregates usage statistics for Claude Code with customizable time periods and data types.