A curated list of articles from data teams building analytics agents.
Recommended first reads · Internal builds · Browse by problem · Related reading · Contribute
- Anthropic — How a team makes self-service analytics work. Skills, metric definitions, documentation, and evaluations (Jun 2026), including how the team keeps them current as data models change.
- OpenAI — Inside an in-house data agent. The architecture behind context preparation, retrieval, permissions, and evaluations (Jan 2026).
- Ramp — Building one around an existing dbt project. SQL and Jinja examples for context tables and domain documentation (Oct 2025).
Read an article directly from the right column, or click a company for its full reading notes. Dates are publication dates; stacks reflect what the authors used at the time.
- Giving the agent context: Ramp, Gorgias, AngelList, Astronomer, Bitrise, BlaBlaCar, Cloudflare, DoorDash, incident.io, Meta, OpenAI, Qonto, Checkr, KOHO, LangChain, Okta, Red Hat, Webflow.
- Testing answers and model changes: Gorgias, Anthropic, ClickHouse, DoorDash, Ramp, LinkedIn, Meta, Uber, Checkr, KOHO, Webflow.
- Fixing mistakes and keeping context current: Bitrise, DoorDash, GitHub, Gorgias, incident.io, Meta, Replit, Astronomer, Checkr, LangChain, Okta, Red Hat, Anthropic, OpenAI.
- Reducing cost and latency: Cloudflare, DoorDash, Gorgias, Vercel.
- Getting people to use it: Alan, Astronomer, Bitrise, Faire, incident.io, Ramp, LinkedIn, Meta, Qonto, KOHO, LangChain, Red Hat, Anthropic, OpenAI.
Product engineering, technical studies, benchmarks, and conference recaps that complement the internal builds.
-
Cube: Building an Agentic Analytics Harness ·
Vendor implementation. Errors, result limits, and permission-aware tools. Stack and notes. -
Omni: Benchmarking Omni’s agentic analytics harness ·
Vendor implementation. Testing quality, latency, and cost on a vendor’s analytics workload. Stack and notes. -
Cassis: Your data context should be debuggable ·
Vendor implementation. Organizing context as a tree of business domains so a missed rule can be traced to the wrong turn that skipped it. Stack and notes. -
Cassis: How to assemble context for analytics agents from your existing assets ·
Vendor implementation. Building the first context from existing data assets, with an open-source bootstrap kit. Stack and notes. -
Cassis: Context engineering for analytics agents ·
Vendor implementation. How we structure tables, metrics, and business rules so an agent can find what it needs. Stack and notes. -
Lyft: From Day 1 to Production ·
Production engineering. Taking an existing analytics agent from a Streamlit prototype to an authenticated, observable internal application. Stack and notes. -
ClickHouse: The Agentic Analytics Benchmark ·
Benchmark. A reusable harness built from 201 real questions, with model-jury scoring, cost and latency measurement, and contamination checks. Code · Notes. -
Benchouse: The Analytics Agent Benchmark ·
Benchmark. Compare accuracy, completeness, restraint, and cost per question across analytics agents. Notes. -
Cassis: Analytics agents at dbt Summit 2026 ·
Conference recap. Where the summit’s 40 sessions on analytics agents and context agreed, where teams still differ, and what the talks left open. Notes. -
Checkr: How Checkr cut an hours-long metrics investigation to minutes ·
Talk write-up. dbt Labs write-up of a dbt Summit 2026 talk. Evals whose failures an agent clusters, drafts a fix for from existing docs, and re-tests before an expert reviews the merge request. Stack and notes. -
KOHO: How KOHO built a Claude-powered self-serve analytics tool ·
Talk write-up. dbt Labs write-up of a dbt Summit 2026 talk. A Claude skill limited to certified dbt marts, tested by an "eval council" of about 375 real analyst questions. Stack and notes. -
LangChain: How LangChain grew self-serve analytics questions 40x ·
Talk write-up. dbt Labs write-up of a dbt Summit 2026 talk. Layered context in Hex: dbt descriptions rewritten for agents, workspace guides kept in the dbt repo, and selective endorsements. Stack and notes. -
Okta: What Okta learned governing context for AI agents at scale ·
Talk write-up. dbt Labs write-up of a dbt Summit 2026 talk. Moving agent context into dbt behind pull requests so Claude and Snowflake Cortex read the same rules. Stack and notes.
Know a team that has written up its internal build? Open an issue or pull request with the link and a sentence about what makes it useful. See the contribution guide for the format. Corrections and broken-link reports are welcome too.