3. Data, Science & AI

Enables semantic search and Dataview-style queries within Obsidian vaults for finding notes by meaning, metadata, and related content.

Automates parallel execution of Claude agent workflows in response to specific user prompts for multi-agent task processing.

Reports AI-detected uncertainties, conflicts, and mistakes to enhance system reliability and user experience.

Handles inter-agent messaging for AI Maestro, enabling message sending, receiving, and inbox management between AI agents.

Proactively searches conversation history to retrieve context, decisions, and prior discussions before processing new user instructions, enhancing contextual awareness in AI interactions.

Queries SRA/GEO datasets, retrieves publications, and analyzes genomics metadata including accession conversion and scRNA-seq identification.

Automates personal knowledge management through entity-based knowledge graphs, extracting facts and synthesizing weekly summaries for structured information retrieval.

Enables context continuity by persisting user session history for seamless conversation flow across multiple interactions.

Distributed git-backed issue tracker for AI agents to manage tasks, decisions, and progress across LLM sessions using Beads (bd).

Comprehensive guide for FinLab quantitative trading package, covering Taiwan stock data analysis, strategy development, and backtesting workflows.

Transforms vague coding requests into structured prompts for Claude Code in Replit, improving AI agent performance through test-driven workflows and task decomposition.

Calculates accurate training costs for Tinker LLM fine-tuning by tokenizing datasets with the correct model tokenizer and providing cost estimates.

Guides data visualization design using Stanford CS448B principles, covering chart selection, encodings, D3.js implementation, and critique.

Implementation guide for vector databases in AI/ML applications, covering RAG, semantic search, and embedding strategies across multiple platforms.

Builds real-time event streaming and data pipelines using Kafka, Flink, and stream processing frameworks for microservices and data integration.

Optimizes vector embeddings for RAG systems via model selection, chunking, caching, and performance tuning to enhance semantic search efficiency.

Builds data pipelines, feature stores, and embedding generation for AI/ML systems, including RAG and feature serving.

Automates LLM system evaluation using metrics, LLM-as-judge, and benchmarks for prompt quality, RAG validation, safety, and model comparison.

Optimizes SQL query performance via EXPLAIN analysis, indexing strategies, and query rewriting for PostgreSQL, MySQL, and SQL Server databases.

Creates interactive business intelligence dashboards with data visualization, KPI tracking, and real-time analytics for executive reporting and monitoring systems.

Deploys LLM and ML models for production inference using specialized tools like vLLM, BentoML, and Triton, optimizing performance and scalability.

Enables efficient data ingestion from cloud storage, APIs, and files into databases for ETL pipeline construction.

Optimizes LLM prompt engineering with zero-shot, few-shot, and chain-of-thought techniques for reliable RAG, AI agents, and cost-effective applications.

Builds AI chat interfaces with streaming, context management, and multi-modal support for ChatGPT-style applications and code copilots.