3. Data, Science & AI

Generates short, URL-friendly public identifiers using nanoid for PostgreSQL databases, while recommending UUID for internal references like auth.users.

Configures Whisper model parameters for speech transcription, enabling adjustments to model size, decoding, and accuracy-speed trade-offs.

Adjusts silence detection and voice activity settings in WaveCap for optimal audio processing, including chunk boundaries, clipping reduction, and latency-accuracy trade-offs.

Configures LLM-based transcription correction settings including model selection, prompt tuning, and Apple Silicon optimization for improved accuracy.

Manages database schema evolution, row-level security policies, and data transformation workflows for data engineering pipelines.

Detects document source type (Zoom, Slack, email, etc.) by analyzing content patterns, enabling automated intake and classification.

Builds LLM applications with LangChain, implementing agents and tool integration patterns for AI workflows.

Generates images using Google Gemini's image generation API via Python CLI for AI image creation tasks.

Analyzes software metrics to provide actionable insights for data-driven decision making and continuous improvement in development processes.

Monitors and manages WaveCap audio streams, including status checks, health monitoring, transcription views, and stream resets.

Analyzes startup user behavior and metrics to validate hypotheses using Lean Startup methodologies.

Reference guide for OpenAI model IDs, capabilities, pricing, and endpoints to support model selection and validation in AI projects.

Maintains a dynamic registry of people by extracting and tracking names, roles, teams, and expertise from documents over time.

Builds RAG systems for LLMs using vector databases and semantic search to ground AI responses in external knowledge.

Retrieves patterns and learned rules from past work to inform cross-agent learning, triggered by memory-related user prompts.

Provides expert guidance for designing, debugging, and optimizing production-grade AI systems, including agent architecture, cost management, and orchestration with LangGraph.

Debugs and analyzes LangGraph agent traces to investigate behavior patterns, failures, and latency issues in AI agent workflows.

Enables local data exploration, profiling, and transformation with DuckDB for CSV, Parquet, and JSON files, generating interactive HTML reports.

Extracts specific entries from JSONL files by ID for data inspection, comparison, and analysis of samples.

Manages tri-tier agent memory (hot/warm/cold) with event consolidation and weekly cleanup to optimize AI agent memory usage after task completion.

Enables comprehensive evaluation of LLM applications through automated metrics, human feedback, and benchmarking to assess performance and quality.

Manages tri-tier memory (hot/warm/cold) for AI agents, including event consolidation and weekly cleanup to optimize memory usage after task completion.

Designs PostgreSQL schemas with best practices, including data types, indexing, constraints, and performance patterns.

Optimizes SQL queries and database schemas using indexing strategies and EXPLAIN analysis to eliminate slow queries and enhance performance.