3. Data, Science & AI

Expertly crafts complex SQL queries, optimizes execution plans, and designs normalized schemas using advanced techniques like CTEs and window functions.

Enables temporal data enrichment and state tracking using OPAL's resource datasets and temporal joins for contextual analysis of events and intervals.

Native execution of Fabric patterns for AI content processing including summarization, claim analysis, and threat modeling without external CLI calls.

Deep dive into OPAL's window functions for row-relative calculations, rankings, and moving aggregates, using window() wrapper (not SQL OVER clause).

Builds LLM applications, RAG systems, and AI agent orchestration with vector search and API integrations.

Detects anomalies in time-series data using statistical methods including Z-score, IQR, and moving averages for identifying outliers and unusual patterns in observability metrics.

Filters and searches event datasets (logs) using OPAL with text, regex, field comparisons, and boolean logic support.

Analyzes latency metrics using tdigest to compute percentiles (p50, p95, p99) for SLO tracking from pre-aggregated data.

Converts unstructured text into structured JSON format based on user-specified fields or provided schema, enabling standardized data processing.

Analyzes event datasets over time using OPAL timechart, enabling trend visualization and metric tracking with configurable temporal binning and output columns.

Enriches datasets by mapping IDs to descriptive names using static CSV reference tables, supporting both explicit and implicit lookups without temporal dependencies.

Provides standardized prompt engineering and context engineering principles for AI agents, optimized for clarity and signal-to-noise ratio based on Anthropic best practices.

Aggregates pre-computed metrics (gauge, counter, delta) using OPAL for analyzing request counts, error rates, and resource utilization over time.

Processes interval datasets (start/end timestamps) using OPAL for duration calculations, temporal filtering, and time-based aggregation in traces and pipeline data.

Expertise in Redis database management for caching, data structures, Pub/Sub, Streams, and vector search using Bun.redis client.

Expert guidance for PostgreSQL database operations including schema design, query optimization, JSONB handling, and performance tuning using Bun.sql client.

Guides pattern selection through AI-driven decomposition, exploration, and diverse approach generation for optimal solution discovery.

Extracts structured fields from unstructured log data using OPAL parsing functions for analysis, filtering, and aggregation.

Aggregates and summarizes event log datasets using OPAL statsby, supporting derived columns, grouping, and statistical functions like count, sum, and avg.

Provides OPAL subquery syntax and union operations for combining datasets, time periods, and performing complex data transformations with temporal shifts and grouping.

Real-time dashboard visualizing multi-agent interactions, tool usage, and session flows with live charts and swim lane visualization for Claude Code sessions.

Automates web content extraction from personal sites and blogs to build knowledge bases via crawling and ingestion.

Manages astronomical data assets including star catalogs and galactic backgrounds for the Stapledons Voyage project.

Provides a systematic framework for selecting AI coding tools (Claude Code vs Gemini CLI) based on project context, codebase size, and task complexity to optimize resource allocation.