Browse Skills

Design and implement structured deep learning training workflows using PyTorch Lightning, covering LightningModule architecture, distributed training strategies, experiment tracking, checkpoint management, and production model deployment.

Expert guidance on designing, building, and querying knowledge graphs using ontology engineering, graph databases, NLP-based construction pipelines, and LLM/GraphRAG integration patterns.

Transforms content to channel-optimal lengths using empirical engagement data for 12+ platforms. Handles truncation-aware restructuring, hook positioning, and format-specific constraints from Reddit to TikTok.

Designs effective prompts and prompt chains for LLM applications. Covers chain-of-thought reasoning, few-shot learning, system prompt architecture, output formatting, prompt evaluation, and reusable template design.

Build production-grade AI-powered features using the Vercel AI SDK with streaming responses, structured outputs, tool calling, and multi-provider support in Next.js applications.

Ensure offline/online feature consistency in ML systems by designing validation pipelines, detecting training-serving skew, and implementing point-in-time correctness guarantees across feature store architectures.

Designs production-grade Retrieval-Augmented Generation pipelines including chunking strategies, embedding model selection, vector database architecture, hybrid search, reranking, and hallucination reduction techniques.

Design AI agent systems with precisely calibrated human oversight, mapping every action to the right autonomy level — from fully automatic to synchronous approval — using established frameworks for trust calibration, escalation policies, and audit trail architecture.

Guides decisions on when and how to fine-tune large language models, covering LoRA/QLoRA, dataset curation, RLHF/DPO alignment, evaluation strategies, and cost-performance tradeoffs for production deployments.

Architect production-grade AI applications with prompt chaining, agent patterns, guardrails, cost optimization, and multimodal capabilities for reliable LLM-powered systems.

Design and implement MLOps platforms covering model serving, feature stores, CI/CD for ML, model monitoring, drift detection, A/B testing, and GPU resource management.

Design and build reliable API integrations in Make (formerly Integromat) using HTTP modules, OAuth flows, pagination handling, and advanced error recovery patterns.

Designs production-grade drift monitoring pipelines for ML models, covering data drift, concept drift, and prediction drift detection with statistical methods, alerting thresholds, and automated response procedures.

Expert in feature store design and operations — real-time feature serving, offline/online consistency, feature pipelines, and ML platform integration with Feast, Tecton, and custom solutions.

Design intelligent model routing systems that optimize cost, latency, and quality across multiple LLM providers. Covers cascading strategies, confidence-based escalation, prompt caching economics, and fallback architectures.

Designs end-to-end computer vision pipelines including model selection (CNN/ViT/YOLO), data augmentation, training strategies, transfer learning, and production deployment with ONNX/TensorRT optimization for cloud and edge.

Expert guidance for building comprehensive observability into LLM-powered applications — covering tracing, cost tracking, quality monitoring, latency analysis, and automated alerting across model providers.

Design and build production-grade n8n automation workflows integrating LLMs, RAG pipelines, and AI agents with robust error handling and cost optimization.

Design and implement anomaly detection systems for tabular data, time series, and streaming environments. Guides algorithm selection, feature engineering, threshold tuning, and production deployment with focus on minimizing false positives.

Systematic optimization of machine learning model hyperparameters using Bayesian optimization, multi-fidelity methods, and distributed search strategies to find optimal configurations efficiently.

Designs 2026-standard RAG systems with agentic retrieval loops, GraphRAG for global reasoning, hybrid search with RRF, supervisor/worker patterns, and evaluation-driven development using RAGAS metrics.

Designs AI agent systems including multi-agent orchestration, tool use patterns, memory architectures, planning algorithms (ReAct, Chain-of-Thought), agent safety guardrails, and production-grade agent frameworks.

Advanced Claude Code orchestration covering subagent strategies, hook engineering, permission architecture, context economy, headless CI pipelines, and session management for power users.

Design and operate low-latency feature delivery systems for production ML, including feature store architecture, online/offline consistency, streaming pipelines, and serving infrastructure optimization.