GCP Data Platform Architect
Bay Area CA
MUST-HAVES Role Scope - Single architect covering both: (1) Salesforce-to-GCP data pipeline build, (2) QUIVER tool deployment
- This is one combined hire, not being split across two roles
Location - Must be Bay Area-based
- Onsite availability required — client is a California government entity
- Standard public-sector vetting/background check should be expected
Employment Type - G5 or G6 level preferred (budget-capped)
- Open to full-time employees and contractors
Technical Capability - Deep GCP data platform expertise: pipeline construction, middleware architecture, job scheduling
- Strong, production-level BigQuery skills: data modeling, performance tuning, partitioning/clustering, slot management, cost optimization
- Hands-on with GCP data stack: Dataflow (Apache Beam), Pub/Sub, Cloud Composer/Airflow, Cloud Storage, Dataform or dbt, Dataplex, Cloud Data Fusion
- Salesforce: functional/admin SFDC knowledge NOT required, but must understand technical constraints — API limits, connector modes (Bulk API, CDC/Platform Events) — and be able to articulate how to design around them
- QUIVER: must be able to work with job definition monitoring, configure autofix capabilities, and validate agent-driven defect fixes
- Strong SQL and Python; Terraform (IaC); CI/CD for data pipelines
- GCP security fundamentals: IAM, VPC Service Controls, encryption, PII handling
- Able to explain end-to-end connectivity architecture (SFDC → GCP) clearly and in own words — this is the key screening test to verify hands-on experience
FULL JD Google Data Platform Architect — SFDC-to-GCP Modernization & QUIVER Implementation Experience: 10+ years in data engineering/platform architecture, with 4+ years hands-on GCP experience Location: Onsite
Role Summary We are looking for a Google Data Platform Architect to own the technical design and delivery of a Salesforce-to-GCP data pipeline program, and to lead the deployment of QUIVER, Tenarai's intelligent DataOps accelerator, on top of the GCP data platform.
This is a dual-mandate role. The first mandate is to land Salesforce CRM data into BigQuery reliably, securely, and at a predictable cost. The second is to make that platform observable, governed, self-healing, and cost-transparent by implementing QUIVER against the GCP stack — so the client inherits an operating model, not just a set of pipelines.
This role is expected to be hands-on. T
hey will write code, run design reviews with client architects, and be accountable for the platform working in production. Key Responsibilities
Workstream 1 — Salesforce to GCP Data Pipeline
- Own the end-to-end target architecture: Salesforce source objects → landing → curated → consumption layers in BigQuery, following a medallion (bronze/silver/gold) pattern.
- Select and justify the ingestion approach across the available options — Salesforce Bulk API 2.0, Change Data Capture and Platform Events, Cloud Data Fusion Salesforce connector, Datastream, Salesforce Data Cloud zero-copy sharing, or a third-party CDC tool — and document the trade-offs on latency, cost, API limit consumption, and operational burden.
- Design incremental and full-refresh strategies for standard and custom SFDC objects, including handling of formula fields, deleted/archived records, soft deletes, and Salesforce governor and API limits.
- Build streaming paths using Pub/Sub and Dataflow where near-real-time is justified; batch paths using Cloud Storage staging and BigQuery load jobs where it is not.
- Orchestrate with Cloud Composer (Airflow 2.x); implement transformation logic in Dataform or dbt with version control, CI/CD, and environment promotion.
- Model the curated layer for downstream consumption in Looker and the client's reporting estate; define partitioning, clustering, and materialization strategy for query performance and cost.
- Implement security and compliance controls: IAM role design, VPC Service Controls, CMEK, Secret Manager, and Cloud DLP for PII in CRM data.
- Own BigQuery cost architecture — editions and slot reservations vs. on-demand, storage tiering, and workload isolation.
- Lead data reconciliation between Salesforce and BigQuery, and define the acceptance criteria the client signs off against.
Workstream 2 — QUIVER Implementation
- Deploy QUIVER as the centralized operations cockpit across the GCP estate, without replacing or duplicating tooling the client already owns.
- Instrument the four QUIVER pillars against GCP-native telemetry.
- Implement predictive alerting: consumption trend warnings and pattern-based failure prediction from historical log patterns and runtime overflow — prevention, not just post-failure retry.
- Integrate QUIVER alerting and remediation workflows into the client's ITSM and change management process.
- Define and hand over the run book, support model, and operating cadence.
Cross-cutting
- Run architecture review board sessions and secure client-side security, privacy, and architecture approvals.
- Produce the artifacts the engagement is judged on: target-state architecture, source-to-target mappings, data contracts, non-functional requirements, and cutover/runbook documentation.
- Mentor engineers on the delivery pod and enforce engineering standards through code review.
- Support pre-sales and account growth with solution shaping and estimation when asked.
Required Skills and Experience
- 10+ years in data engineering or data platform architecture, including 4+ years designing and delivering on Google Cloud.
- Deep, production-level BigQuery expertise: data modeling, performance tuning, partitioning and clustering, slot and reservation management, cost optimization.
- Hands-on with the GCP data stack — Dataflow (Apache Beam), Pub/Sub, Cloud Composer/Airflow, Cloud Storage, Dataform or dbt, Dataplex, Cloud Data Fusion.
- Demonstrated experience integrating Salesforce as a data source at enterprise scale, including working knowledge of the Salesforce data model, Bulk API, CDC/Platform Events, and API governor limits.
- Strong SQL and Python; infrastructure-as-code with Terraform; CI/CD for data pipelines.
- Practical experience with data quality, observability, and lineage tooling, and with designing operational alerting that engineers actually act on.
- Cloud security fundamentals on GCP: IAM, VPC Service Controls, encryption, PII handling.
- Proven ability to run technical conversations with enterprise client architects and security teams, and to defend design decisions.
Preferred
- Google Cloud Professional Data Engineer or Professional Cloud Architect certification.
- Experience implementing a DataOps, observability, or FinOps platform (e.g. Monte Carlo, Collibra, Ataccama, Atlan, Unravel)
- Prior multi-cloud exposure (Azure Data Factory, Databricks, AWS) for hybrid estates.
- Looker or LookML modeling experience.
- Consulting or systems integrator background with client-facing delivery accountability.
