Original job description
Join Hotspex Media!
🙌 #1 Ranked Media Buying and Planning Agency on Clutch.co
🔥 Finalist 'Best AI Tool', 2024 Digiday Technology Awards
🎉 Hybrid Work Model (1 Day in Office / Week)
🥇 Winner of Waterstone Canada's Most Admired Corporate Cultures
Reports to: Director of AI & ML
Location: Hybrid with the option for Remote if Outside Greater Toronto Area (must be legally authorized to work in and based in Canada)
Team: Small, high-autonomy team with direct access to leadership.
Impact: Owns design, build, operation of Hotspex's data transformation and storage layer.
About the Role & Mission
Connect data across BigQuery, Postgres, and Airtable; expose clean datasets to AI, Workflow, Analytics consumers
Build and maintain dbt models transforming marketing platform data into conformed dimensional schemas (Kimball facts/dimensions)
Own SQL surface: queries, stored procedures, views, materialized views, scheduled routines
Optimize warehouse performance and cost: query tuning, partitioning, clustering, incremental models
Orchestrate pipelines with Airflow or similar
Core Competencies
SQL Engineering: Writes, tunes, maintains complex SQL across BigQuery and Postgres
Stored Procedures & Routines: Designs and owns stored procedures, scripted procedures, UDFs, scheduled jobs
dbt / Transformation Modeling: Builds and maintains dbt models with tests, docs, incremental patterns
Pipeline Orchestration: Schedules and monitors pipelines via Airflow or similar
Cross-Functional Partnership: Delivers consumable data products for AI, Workflow, Analytics
Job Specific Competencies
Advanced SQL: Complex joins, window functions, CTEs, query optimization, execution plans on BigQuery and Postgres
Stored Procedures & Routines: Production stored procedures, scripted procedures (BigQuery scripting / PL/pgSQL), UDFs, scheduled queries with error handling, idempotency, observability
dbt Modeling: Sources, staging, intermediate, marts; tests; documentation; incremental strategies; macros
Pipeline Orchestration: Airflow, Dagster, Prefect, or equivalent
Data Modeling: Kimball facts/dimensions, slowly changing dimensions, conformed schemas
Warehouse Optimization: Partitioning, clustering, materialized views, cost tuning on BigQuery
Airtable Integration: Schema mapping, sync patterns, base-as-source
Job ResponsibilitiesConnect & Optimize Data
Own connectivity between BigQuery, Postgres, and Airtable; ensure consumers (AI, Workflow, Analytics) get the schema they need
Refactor ad-hoc SQL into versioned, tested, documented routines
Optimize cost and performance: partitioning, clustering, materialization
Detect and fix performance regressions before downstream impact
SQL & Stored Procedure Ownership
Own every production stored procedure, scripted procedure, scheduled query across BigQuery and Postgres
Author new stored procedures for batch transforms, reporting routines, AI/ML feature prep
Maintain stored-procedure inventory with ownership, dependencies, runbooks
dbt Model Build & Maintenance
Design schemas and write dbt models transforming marketing platform data (Google Ads, Meta, LinkedIn, etc.) into conformed dimensional schemas
Implement dbt tests (uniqueness, not-null, referential integrity, custom rules) on every production model
Maintain incremental models for high-volume tables; tune for cost and freshness
Own dbt documentation and lineage
Pipeline Orchestration
Schedule, monitor, and version pipelines in Airflow or similar
Alert routing, retry policy, backfill patterns
Coordinate with Workflow Eng on hand-off points between n8n and orchestrated data pipelines
Data Quality, Monitoring & Reliability
Implement automated tests (dbt tests, freshness checks, row-count anomaly detection)
Detect and acknowledge data quality incidents within 1 business hour (SLA)
Author runbooks for common failure modes
Track and reduce incident frequency; report trends quarterly
Cross-Functional Partnership
Partner with Workflow Automation Engineer on ingestion contracts: landing schemas, refresh patterns
Partner with Junior AI Engineer on data needs for RAG, embeddings, AI services: feature tables, serving views
Translate PM/CS and Product requirements into dimensional models
Owns: SQL design, stored procedure logic, transformation modeling, performance choices
Does not own: automation logic (Workflow Eng), AI service code (Jr AI Eng), client-facing strategy
Documentation & Knowledge
Use Claude Code for stored procedure docs, model READMEs, schema references
Version-controlled repos, clean Markdown, proper Git hygiene
Document data contracts: ingestion → transformation → consumption
Continuous Improvement
Use AI tooling (Claude Code, Cursor) to accelerate SQL authoring, refactoring, documentation
Track and report query cost reduction and model freshness improvement quarterly
Resolve categories of technical debt: consolidating duplicated SQL, retiring shadow tables
Explicitly Out of Scope
n8n automation design and ownership (Workflow Automation Engineer)
Rust service development, RAG pipelines, embedding models (Junior AI Engineer)
Looker dashboard authoring and LookML feature development
Strategic analytics presentations to leadership
ML model engineering, training, prompt engineering as a discipline
Required Qualifications
2+ years data engineering, analytics engineering, or database development
Strong SQL - complex joins, window functions, CTEs, query optimization (must demonstrate)
Hands-on stored procedure experience - production stored procedures (BigQuery scripted procedures, PL/pgSQL, T-SQL, PL/SQL, or equivalent). Non-negotiable.
Working knowledge of dbt (or strong SQL/Git fundamentals to ramp quickly)
Python or other scripting language for data tasks (Java, Scala, TypeScript also acceptable)
Airflow or similar pipeline orchestration experience (Dagster, Prefect, dbt Cloud schedules, Cloud Composer)
Dimensional modeling fundamentals - facts, dimensions, grain, conformed schemas
Git fundamentals - branches, PRs, code review participation
Documentation discipline - version-controlled Markdown
Strongly Preferred
BigQuery production experience (partitioning, clustering, scripted procedures, scheduled queries)
Postgres production experience (PL/pgSQL, indexes, query plans)
Airtable production experience (schema design, sync patterns, API integration)
Production dbt experience (Cloud or Core)
Marketing/advertising data sources (Google Ads, Meta, LinkedIn)
AI tooling (Claude Code, Cursor, ChatGPT) as daily accelerator
Nice to Have
Looker / LookML exposure (consumer-side; not required to own)
n8n or other workflow orchestrators
RAG / vector search data prep
Agency, media, or analytics domain
Technology Stack
Languages: SQL (advanced), Python (or equivalent), optionally JavaScript for dbt/BigQuery UDFs
Data: BigQuery, Postgres, Airtable, dbt, Redis (cache awareness)
Orchestration: Airflow (or Dagster, Prefect, Cloud Composer), dbt Cloud Run
Integration consumer-side: n8n
Cloud: GCP
Observability: Cloud Monitoring, Looker (consumer-side)
Tools: Linear, GitHub, Claude Code, Cursor
What this Role is NOT
Not a data analyst - no ad-hoc analysis, dashboarding, stakeholder reporting
Not analytics engineering / dashboards - LookML and Looker dashboards not owned
Not workflow automation - n8n belongs to Workflow Automation Engineer
Not ML / AI engineering - model development belongs to AI team
Not a DBA - no infrastructure provisioning or cluster management
Our Values:
We know our people are what allows us to achieve all that we do and that’s why it’s important that everyone we bring onto our team lives our values with us.
🦁 Courage
1️⃣ One Team
💪 Resilience
🚀 Empowerment
Hotspex Media Inc. is an equal opportunity employer and values diversity in its workforce. Due to the large volume of applications received, Hotspex Media may, from time to time, use artificial intelligence to optimize screening efforts.
