Platform Architect
No salary listed
Expert
San Francisco, CA, USA
In Person
DevOps & InfrastructurePlatform Engineering
About the job
Requirements
- At least 10 years of experience in large-scale distributed systems or platform infrastructure.
- Deep expertise in build systems and dependency management at scale, such as Bazel, Buck, Pants, or Nx, involving 1,000 or more packages.
- Production experience with reproducible build engineering, including content-addressable storage, hermetic builds, deterministic continuous integration, and artifact provenance such as SLSA.
- Experience building dependency graph tooling for static and runtime dependency analysis, impact computation, and selective testing.
- Strong background in semantic versioning governance across large service fleets, including version resolution, compatibility constraints, and lockfile-style pinning.
- Experience with API contract evolution, including protobuf/gRPC versioning, schema evolution strategies, and contract testing.
- Experience with multi-region, multi-tenant platform architectures serving regulated workloads.
- A track record of migrating monolithic systems to modular architectures without business disruption.
Responsibilities
- Own the version engineering and module governance architecture to ensure reproducible deployments, computable change blast radius, and known-good site compositions.
- Design a version set model in which each site is a locked, tested combination of module versions, with version compatibility matrices and automatic constraint solving.
- Architect reproducible build pipelines so modules produce byte-identical artifacts regardless of build time or location, using hermetic build environments, content-addressable artifact storage, and SLSA Level 3 or higher provenance.
- Build and maintain a live dependency graph spanning more than 2,000 services, combining static code, API, and schema analysis with runtime traffic-weighted call patterns, transitive dependency resolution, and automatic conflict detection.
- Design a monorepo and package ownership governance model, including module boundaries using domain-driven design, CODEOWNERS enforcement, cross-module change protocols, and module-boundary tooling such as lint rules, continuous integration checks, and architectural fitness functions.
- Architect change impact analysis systems that compute affected modules, sites, and tests for each pull request; run only dependency-graph-proven relevant tests; and score change confidence using blast radius, coverage, and stability to produce a risk grade.
- Design a GitOps-driven multi-site deployment model with site configuration repositories declaring module versions and compliance parameters, and automated strategies for patch auto-deployment, minor releases with review, major releases with migration plans, and security force-pushes.
- Architect a four-layer testing framework comprising module self-tests, country-parameterized platform compliance tests, cross-module integration tests, and site-specific tests, with automated coverage enforcement that blocks deployments when gaps are detected.
Desired Qualifications
- Experience with Amazon's internal VersionSet, Brazil build system, or Apollo deployment, or similar systems at Google such as Blaze/Bazel or Meta such as Buck/Configerator.
- Background at Stripe or Adyen, involving multi-country financial platforms with compliance parameterization.
About the company
Embedding VC is an early-stage venture fund that backs Generative AI startups. It provides seed to early-stage capital and operational guidance, leveraging a team of current and former builders and operators who can help portfolio companies grow. The fund’s approach centers on working closely with founders and using hands-on experience to scale AI projects, including those aiming toward AGI, during a period described as a new computing super cycle driven by AI innovations. Unlike funds that rely mainly on financial leverage, Embedding VC differentiates itself through its deep operator network, practical support, and focus on and commitment to building capable, scalable AI businesses. The overall goal is to support ambitious AI founders to advance the development and deployment of AI technologies and capture value in the expanding AI ecosystem.
Company Size
11-50
Company Stage
N/A
Total Funding
N/A
Headquarters
Menlo Park, California
Founded
2023
Get referred to Embedding VC
See people who can refer or advise you
Simplify's Take
What believers are saying
- Embedding VC's July 2026 site update confirms active positioning in Menlo Park.
- Roger Luo reports over forty startup investments, including three IPOs and nine exits.
- The fund claims portfolio support wins, including AppViewX integration and Bessemer placements.
What critics are saying
- Portfolio concentration in frontier AI ties returns to a crowded, winner-take-most market.
- No recent public fundraising or new flagship exits reduces evidence of momentum.
- Founder-led branding around Roger Luo creates key-person risk if he steps back.
What makes Embedding VC unique
- Roger Luo markets Embedding VC as an operator-led Generative AI fund.
- The firm targets pre-seed and seed AI founders, not broad venture.
- It emphasizes hands-on startup building through its Founder-in-Residence program.
Help us improve and share your feedback! Did you find this helpful?
Benefits
- Health Insurance
- Company Equity