Data skills

Generates read-only MongoDB queries and aggregation pipelines from natural language, using the collection schema.

Helps optimise slow MongoDB queries and choose the right indexes for them.

Covers MongoDB schema design patterns and anti-patterns, including migrations from SQL.

Guides implementing Atlas Search, Vector Search and Hybrid Search on MongoDB.

Generates anywidget components, which add custom interactive widgets to marimo notebooks.

Converts a Jupyter notebook (.ipynb) into a marimo notebook (.py).

Writes a marimo notebook as a Python file in the correct format.

Converts a Streamlit app into a marimo notebook, mapping its widgets and layout to marimo equivalents.

Covers Redis client connections: pooling, pipelining, client-side caching, slow commands to avoid and timeouts.

Helps choose the right Redis data structure for a data model and keep key names consistent.

Covers Redis Search: index schema design, field types, query syntax and vector similarity search.

Covers semantic caching of LLM responses with Redis LangCache, including similarity thresholds.

Helps choose and create the right Neon branch type for testing, migrations and isolated development.

Guides work with Postgres on Neon: connections, pooling, migrations, branching, autoscaling and read replicas.

Builds or updates a PostHog dashboard from a set of insights through MCP.

Guides the creation of a PostHog experiment: hypothesis, rollout, bucketing, analytics and running time.

Creates and launches PostHog surveys through MCP, including NPS and CSAT popovers and hosted feedback forms.

Adds PostHog product analytics events so that meaningful user actions are captured.