Bibhas Debnath

Bibhas Debnath

@iambibhas

Decentralized Ownership, Centralized Control: Building a Self-Healing Data Platform

Submitted Jun 25, 2026

Brief abstract

Bringing together all the org data by following a data mesh architecture in a decentralized manner sounds great at start, but the approach quickly falls on its face because of strict data governance policies around cross-team data access. Getting governance team approve access request for hundreds of teams get real painful and slow immediately, and the data product teams start falling back to what we call “Shadow IT” - personal google sheets, excel files, CSV files, ad-hoc reports/dashboards etc, shared without any oversight.

This session breaks down how we solved this paradigm at scale by building an internal data platform where compliance is declarative and infrastructure is self-healing. By leveraging a GitOps workflow, we shifted compliance and access control from an afterthought of manual checklists directly into the version control repository. And with the Openshift operator pattern, we made the platform self healing that always adjusts itself to provide the right access to the right group of people.


Key Takeaways

  • How to implement the Reconciler Pattern to continuously monitor and self-heal cloud data infrastructure against configuration drift.
  • How to architect a GitOps-driven compliance pipeline where Git commits and PR approvals serve as an automated, immutable audit trail.
  • Practical strategies for scaling Declarative RBAC, transferring access approval rights entirely to data product teams while maintaining centralized security guardrails and audit trails.

Audience

  • Data Engineers and Data Architects scaling distributed data platforms.
  • DevOps, SRE, and Platform Engineers focused on GitOps workflows and state reconciliation.
  • Security & Infrastructure engineering leaders trying to eradicate Shadow IT without crippling engineering velocity.

Bio

I am Bibhas Debnath, a seasoned engineer with experience in various stages of product lifecycle for over a decade. I am currently working at Red Hat, blending SQL with Kubernetes (Openshift).


Draft slides
https://docs.google.com/presentation/d/1zWU_H6HJyP06xLDvtKcmBPQMckATj8M7mj27DBTCh30/edit?usp=sharing


Video pitch

<missing>
I’ll share this on 26th June. Not in a suitable position to make it right now.

Comments

{{ gettext('Login to leave a comment') }}

{{ gettext('Post a comment…') }}
{{ gettext('New comment') }}
{{ formTitle }}

{{ errorMsg }}

{{ gettext('No comments posted yet') }}

Hosted by

Jumpstart better data engineering and AI futures