Skills

Defines versioned data contracts between producers and consumers, with ownership, schemas, quality expectations, compatibility rules, and operational change management.

Designs trustworthy product analytics events, properties, identity rules, validation, privacy boundaries, dashboards, and governance for product and growth decisions.

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

Inspects messy tabular data and produces a repeatable cleaning plan plus code. Detects and fixes common issues: inconsistent types, duplicate rows, missing values, malformed dates, mixed encodings, whitespace and casing problems, and outliers. Outputs pandas or Polars code, a summary of changes, and a validation checklist.

Generates and refactors dbt models for analytics engineering. Writes staging, intermediate, and mart models following layered conventions, adds schema.yml tests and descriptions, applies incremental and materialization strategies, and structures sources and refs correctly. Produces SQL plus YAML that fits dbt best practices and is ready to run.

Turns analysis results into a clear narrative for a specific audience. Selects the key message, orders findings for impact, recommends the right chart for each point, writes plain-language takeaways, and frames actionable recommendations. Helps analysts move from raw numbers to a memo or slide narrative executives can act on.

Explains what a SQL query does in plain language and how it executes. Breaks down joins, subqueries, CTEs, and window functions step by step, describes the result set, reads EXPLAIN/EXPLAIN ANALYZE output to identify slow scans and missing indexes, and flags correctness pitfalls. Helps developers understand, review, and trust unfamiliar SQL.

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.

Defines data contracts between producers and consumers to prevent breaking changes in pipelines. Covers schema definitions, semantic types, freshness and quality SLAs, ownership, versioning, and backward/forward compatibility rules. Generates contract specs (e.g., ODCS-style) and CI checks that fail builds when a producer violates the contract.

Helps design and implement features for machine learning models from raw tabular, time-series, or text data. Suggests transformations, encodings, aggregations, and leakage-safe splits, explains the rationale, and generates reproducible feature pipeline code with validation.

Designs data quality checks for tables, pipelines, and warehouses. Generates expectation suites covering schema conformance, null and uniqueness constraints, referential integrity, freshness, and statistical drift, then wires them into pipelines so bad data is caught before it reaches dashboards or models.

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 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.

Generates realistic synthetic datasets that preserve the statistical properties and relationships of source data without exposing real records. Covers schema-aware generation, correlated and time-series fields, class balancing for ML training, and constraint preservation, with code for tools like SDV, Faker, or custom generators.

Builds retention and behavioral cohort analyses from event or transaction data. Defines cohorts by acquisition date or attributes, computes retention and churn curves, generates the SQL or pandas code to produce cohort tables, and interprets the results into actionable insights about engagement and lifecycle.

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.

Turns raw data and natural-language requests into clear, well-labeled charts and the code to render them. Recommends the right chart type for the data and message, handles aggregation and formatting, and outputs production-ready visualizations using libraries like Matplotlib, Plotly, Vega-Lite, or Chart.js with accessible color palettes.

Classifies the sentiment and emotional tone of text — reviews, support tickets, social posts, and survey responses. Supports document-level and aspect-based sentiment, returns confidence scores and representative quotes, and aggregates trends across large batches with themes and actionable insights.

Extracts structured data from PDFs and scanned documents — invoices, receipts, forms, contracts, reports, and tables. Returns clean, typed output (JSON, CSV, or Markdown tables), handles multi-page layouts and nested tables, and flags low-confidence fields for review. Uses vision-capable models for image-based and scanned PDFs.

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.

Builds and evaluates time series forecasting models using statistical and ML approaches. Supports ARIMA, Prophet, LSTM, Transformer-based models, and foundation models like TimesFM. Handles seasonality detection, trend decomposition, anomaly detection, multi-step forecasting, and backtesting with proper train/test splits for financial, IoT, and scientific time series data.