Krishna Dey Work · Platform priority
Multi-Workspace & Explicit Tenant Ownership
A platform program to add explicit tenant identifiers to data models that currently infer workspace ownership through joins, and to move authentication from a one-user-one-workspace assumption to one identity across many workspaces.
Engineering Manager, Platform
Ongoing platform program.
Challenge
Several older systems assume one user maps to one workspace, and some data models infer tenant ownership through relationships rather than carrying an explicit identifier. That limits users who belong to multiple workspaces, multi-region or multi-entity customers, safe workspace deletion, and tenant auditing.
Approach
I connected workspace deletion, tenant mapping, isolation, auditing, access control, and multi-workspace identity into one platform sequence rather than separate fixes. The plan defines a common tenant-key schema, updates application writes to attach it, backfills historical rows, migrates reads to the explicit key, and only then enforces non-null constraints. On the identity side, the proposed model issues a short-lived login token after first authentication, returns the workspaces available to that identity, and requires an explicit exchange (including SSO where required) before granting workspace-scoped access.
Outcome
This is active 2026 platform work. Compatibility for existing single-workspace users, strict prevention of cross-workspace access, and administrative control are part of the stated success criteria alongside performance and security.
What it taught
Explicit tenant ownership is foundational infrastructure: it is what makes safe deletion, auditing, and future data partitioning possible, but it has to be sequenced carefully against a system that has run on inferred ownership for years.
More projects
All projectsNative integration and platform work spanning CRM, procurement, and collaboration systems, with emphasis on reusable primitives, metadata-aware design, and integration surfaces that stay coherent as the catalog expands.
ClickWrap captures recorded acceptance of terms in-product — a checkbox, link, or agreement viewer on a website, app, or signup flow — without sending the agreement through a full negotiation and e-sign ceremony. We explored that product shape as a 2023 hackathon with Sanket Mishra and Punit Batra; another team later shipped it. The shipped product has had more than $1M in revenue impact.
An Amazon EFN chapter focused on internal fulfillment systems, requirements engineering, data pipelines, marketplace-launch automation, and high-scale inventory services where correctness and latency both mattered.