Overview·Topics·Wrap-Up

BUILD SPRINT REVIEW

JUNE 2025

July 2025

Overview·Topics·Wrap-Up

OVERVIEW

WHAT WE WORKED ON

Overview·Topics·Wrap-Up

SPRINT SCOPE

  • 4 active topics in the June BUILD sprint
  • Focus: developer productivity, CI efficiency, AI products, and automated maintenance
  • Each topic owner presents their progress after this overview

Not in this review: #4 SSH → 1Password (back to backlog) · #30 Tasty Urban (finalization in July)

Gemma Analytics | BUILD Sprint Review — June 2025
Overview·Topics·Wrap-Up

TOPICS AT A GLANCE

#1
AIRFLOW3 TEMPLATE REPO
Clean, reliable starting point for every new Airflow project
#8
DBT SLIM CI
Run only what changed — faster, cheaper CI pipelines
#18
BI CHATBOT PROTOTYPE
Conversational analytics via Snowflake Cortex
#27
GEMMA LENS AGENT V1
Automated tech debt, GDPR, and security scanning
Gemma Analytics | BUILD Sprint Review — June 2025
Overview·Topics·Wrap-Up

TOPICS

SPRINT PROGRESS

Overview·Topics·Wrap-Up

#1 — AIRFLOW3 TEMPLATE REPO

Owner: Dima

Problem: The shared repo is inconsistent and unreliable as a starting point — engineers need tribal knowledge to get going.

Goal: Clean it up so any engineer can use it straight away, no context needed.

2–4h
Saved per new project
Gemma Analytics | BUILD Sprint Review — June 2025
Overview·Topics·Wrap-Up

#8 — DBT SLIM CI

Owner: Matthew

Problem: CI pipelines run the full test suite every time — slow and costly.

Goal: Pilot Slim CI on one client project to limit runs to changed models only. Package as a reusable Gemma best practice.

40–70%
Target runtime reduction
Gemma Analytics | BUILD Sprint Review — June 2025
Overview·Topics·Wrap-Up

#18 — BI CHATBOT PROTOTYPE

Owner: Cornelius

Problem: Business users depend on analysts for every data question.

Goal: Build a conversational layer on top of the data warehouse (Snowflake Cortex) so users can ask questions in plain language and get governed answers.

🎯 Aimed at becoming a pitchable AI service offering for Gemma
Gemma Analytics | BUILD Sprint Review — June 2025
Overview·Topics·Wrap-Up

#27 — GEMMA LENS MAINTENANCE AGENT V1

Owner: Max

Problem: Technical debt, GDPR risks, and security gaps across client repos are hard to detect at scale manually.

Goal: Automated agent that scans client repos and flags issues. V1 live for at least one pilot client.

🎯 Positioning Gemma Lens as a retainer-justifying, differentiating product
Gemma Analytics | BUILD Sprint Review — June 2025
Overview·Topics·Wrap-Up

WRAP-UP

DISCUSSION & NEXT STEPS

Overview·Topics·Wrap-Up

WHAT'S NEXT

  • Feedback round — what went well, what needs adjusting
  • July priorities — carry-overs, new items, #30 finalization
  • Dependencies — anything blocking progress across topics
Gemma Analytics | BUILD Sprint Review — June 2025
Overview·Topics·Wrap-Up

THANK YOU

LET'S DISCUSS

Welcome to the first BUILD sprint review. This deck gives a quick overview of each topic — the owners will then walk us through the details.

Quick framing before we dive into the individual topics. Two items dropped out of this sprint's scope.

Four topics, each addressing a different dimension: developer experience, CI efficiency, AI product, and automated maintenance.

Dima walks us through the changes. Key question: is the template usable as-is today, or are there remaining blockers?

Matthew shares pilot results. Key questions: which client was this tested on, and what's needed to make it a standard across projects?

Cornelius demos the prototype. Key questions: what's the accuracy like, and what does the path to a client pilot look like?

Max shares v1 status. Key questions: which client is the pilot, what categories of issues does it detect, and what's the false positive rate?

Open floor for discussion. Let's hear from each topic owner: what do you need from the team to keep moving?