Skills

Evaluates retrieval quality for search and RAG systems using grounded test sets, relevance metrics, failure taxonomy, chunking experiments, and actionable remediation.

Designs evaluation harnesses for LLM applications, covering dataset construction, task-specific metrics, LLM-as-judge rubrics with bias controls, and regression gates. Helps teams measure quality, catch regressions across model or prompt changes, and report results with confidence intervals rather than vibes.

Designs recommendation systems end to end: candidate generation, ranking, and re-ranking. Covers collaborative filtering, content-based and embedding retrieval, two-tower models, cold-start strategies, feature stores, offline/online evaluation (NDCG, recall@k), and feedback loops. Produces an architecture and evaluation plan tailored to the product.

Curates high-quality datasets for supervised fine-tuning (SFT) and preference optimization (DPO/RLHF). Covers deduplication, quality filtering, formatting into chat/instruction templates, train/validation splits, label balancing, contamination checks against eval sets, and PII scrubbing. Produces clean, well-documented datasets ready for training.

Designs and implements retrieval-augmented generation (RAG) pipelines end to end. Covers document chunking strategies, embedding model selection, vector store configuration, hybrid and re-ranking retrieval, prompt construction with grounded citations, and evaluation harnesses for measuring retrieval quality and answer faithfulness.

Designs and generates data pipeline configurations for ETL/ELT workflows. Supports Apache Airflow DAGs, dbt models, Spark jobs, and streaming pipelines with Kafka or Flink. Creates data quality checks, schema evolution strategies, and monitoring dashboards for pipeline health.

Analyzes video content to extract insights, describe scenes, identify objects and actions, generate summaries, and create structured annotations. Supports temporal reasoning across frames, scene change detection, and content categorization. Works with educational videos, product demos, surveillance footage, and social media content.

Performs cheminformatics and molecular property analysis using RDKit, PubChem, and ChEMBL. Supports SMILES/InChI parsing, molecular descriptor calculation, drug-likeness filtering (Lipinski, Veber), ADMET prediction, substructure search, and structure-activity relationship (SAR) analysis for drug discovery workflows.

Builds and executes bioinformatics analysis pipelines for genomics, transcriptomics, and proteomics data. Supports single-cell RNA-seq analysis with Scanpy, differential expression with PyDESeq2, sequence alignment, variant calling, gene ontology enrichment, and pathway analysis using KEGG and Reactome databases.

Builds and queries typed knowledge graphs for structured agent memory and composable skills. Creates entities (people, projects, tasks, events, documents), links related objects, enforces constraints, and plans multi-step actions as graph transformations. Enables persistent, queryable memory across agent sessions.

Analyzes SQL queries and database schemas to identify performance bottlenecks and suggest optimizations. Recommends index strategies, query rewrites, denormalization opportunities, and partitioning schemes. Explains EXPLAIN plans and provides before/after comparisons with expected performance improvements.

Designs normalized database schemas from business requirements. Covers entity relationships, indexing strategies, migration planning, and performance considerations. Supports PostgreSQL, MySQL, MongoDB, and other databases with dialect-specific optimizations.