Hi, I’m Julia Shem.
I LOVE solving data problems, whether it’s migrating off a legacy system, building data foundations from scratch, architecting models and semantic layers, helping a team finally trust the numbers in front of them.
Data is messy by nature — tools multiply, sources sprawl, and suddenly nobody agrees on what a number means. What I actually love is building the clean foundation underneath it all: efficient and well-structured.
Most recently as Principal Analytics Engineer at Lifeforce I built the entire data platform end-to-end: ingestion with Fivetran, transformation with dbt (100+ governed models), Snowflake as the warehouse, and Omni as the BI and semantic layer, queried by both teams and AI agents. Before that, I led dbt implementations, built enterprise data governance, and managed data infrastructure across 200+ servers.
Through 8020Analytics.co I work with companies at various stages on end-to-end data builds — helping teams go from scattered, untrusted data to a clean foundation with governed metrics and reliable answers.
I work across a modern data stack built to scale: Fivetran / Airbyte for ingestion, Snowflake / Databricks as your warehouse, dbt for transformation and semantic layer, and Omni / Looker as your AI-forward BI layer. My standard: clean models, proper documentation, and architecture that outlasts the engagement.
If your data is scattered, your metrics mean different things to different people, or your AI tools are giving answers nobody trusts — I’d love to help.
Defining the metrics that matter — cohort retention, LTV, conversion, product engagement — and getting them agreed-on across stakeholders. Because the best data platform in the world means nothing if your teams are still arguing about the numbers.
A modern data stack built to scale with your growth: Fivetran/Airbyte for ingestion, Snowflake/Databricks as your warehouse, dbt for transformation and semantic layer, and Omni/Sigma as yourAI analytics platform. One clean, connected platform.
Centralized data from your operational tools — payments, e-commerce orders, CRM, marketing, etc — into one data warehouse. Everything in one place, ready to model.
Fast, well-documented dbt models your team can extend. Or migration and cleanup of an existing dbt project that’s grown into spaghetti.
Built with governance in mind — PII isolated, access controls in place, and models documented so anyone can audit them.
Including a governed semantic layer so your metrics are defined once and trusted everywhere — by your BI tools, your teams, and your AI agents.
Migrating from traditional BI tools (Tableau, Looker, Mode) to an AI-native analytics layer like Omni or Sigma — where reports are fast, metrics stay consistent, and your data is structured so AI tools return answers you can trust.