Technical Product Management · Data Governance · AI
AI that organizations
can actually trust.
Punica Bhardwaj · TPM · Data Governance . 12+ Years
I believe ungoverned AI isn't a product - it's a liability. With 12+ years spanning technical product management, data governance, and program management at companies like T-Mobile and EY, I've learned that the organizations winning with AI are the ones building governance in from day one.

This portfolio is working proof of that thinking. Every tool here is mapped to a real regulatory framework, built with auditable logic, and ready to hold up under scrutiny.
NIST AI RMF EU AI Act ISO/IEC 42001 Singapore Agentic AI Apache Atlas Alation CSPO SAFe Scrum Master Databricks Data Engineer
See the Tools About Me
🔒
Governance First
AI products without governance structures aren't finished products - they're future incidents waiting to happen. I design governance in from the first sprint, not the last.
📐
Auditable by Design
Every decision a model makes should be traceable. I build systems where risk classification, fairness assessments, and vendor evaluations produce defensible, documented outputs.
🤝
Trust is the Product
The organizations that will win with AI are the ones that earn trust early - with regulators, with customers, with employees. Governance isn't overhead. It's competitive advantage.
Background
12+ Years Building
Products That Last
My career sits at the intersection of technology, data, and governance. I've led product and program initiatives across telecom, consulting, and enterprise data - always with a focus on building systems that scale responsibly.
My foundation is in data governance and information assurance, which shapes how I approach AI: with rigor, with documentation, and with a clear understanding of what happens when things go wrong.
Manager, Technical Product Management
T-Mobile · Bellevue, WA
Led technical product initiatives across data and AI domains at scale. Drove platform decisions balancing speed, compliance, and enterprise governance requirements.
Technology Consultant - Manager
EY US LLP · Seattle, WA
Delivered technology consulting engagements with a focus on data governance, program management, and risk. Worked across regulated industries where accountability and documentation matter.
Education & Credentials
The Foundation
My academic background in information assurance and computer engineering gives me the technical depth to understand AI systems - and the governance vocabulary to make them trustworthy.
M.S. Information Assurance & Cyber Security
Graduate degree focused on security architecture, risk management, and data protection
B.Tech Computer Engineering
Technical foundation in systems, software, and computer science fundamentals
CSPO
Certified Scrum Product Owner - product lifecycle and stakeholder alignment
SAFe Certified Scrum Master
Scaled Agile Framework delivery across large enterprise programs
Databricks Data Engineer
Certified in data engineering, pipelines, and lakehouse architecture
Data Governance Tools
Apache Atlas · Alation · data lineage, cataloging, and metadata management
Governance Prototype Portfolio
AI Governance Tools
Four working prototypes covering the full AI governance lifecycle - intake, runtime, fairness, and vendor risk. Each uses deterministic rule engines mapped to real regulatory frameworks.
PROTOTYPE 01 LIVE
AI Use Case Intake & Risk Classifier
A structured intake tool for product teams submitting AI systems for governance review. Walks teams through a 4-step questionnaire, then classifies risk level and maps it to NIST AI RMF tiers, EU AI Act risk bands, and ISO/IEC 42001 controls.
NIST AI RMF EU AI Act ISO/IEC 42001 Risk Classification Intake Governance
Launch Tool →
PROTOTYPE 02 LIVE
Agent Governance Console
A runtime governance dashboard for autonomous AI agents. Demonstrates permission scoping (allow / block / approval-required), immutable action audit logs with full decision traces, one-click kill switches, and policy override tracking.
Agentic AI Singapore Framework NIST AI RMF Manage Runtime Controls Audit Log
Launch Tool →
PROTOTYPE 03 LIVE
Bias & Fairness Audit Tool
A pre-deployment fairness analysis tool. Upload model prediction data with demographic attributes to compute three industry-standard fairness metrics: Disparate Impact Ratio (EEOC 4/5ths Rule), Equal Opportunity Difference, and Calibration Parity.
Bias Detection Disparate Impact EEOC 4/5ths Rule NIST AI RMF Measure Fairness Metrics
Launch Tool →
PROTOTYPE 04 LIVE
AI Vendor Risk Scorecard
A third-party AI vendor assessment tool covering six weighted governance domains: model transparency, data governance, security, regulatory compliance, incident response, and contractual protections.
Vendor Risk Third-Party AI NIST AI RMF Govern Procurement Contractual Controls
Launch Tool →
AI Product Portfolio
AI Products I've Built
Applied AI tools demonstrating technical product thinking beyond governance - from intelligent workflow automation to decision support systems. More coming soon.
🛂
H1B Job Navigator
The job search tool that immigration status forgot to build. Searches live LinkedIn listings at 35+ H1B-verified companies filtered for TPM, PM, and AI Governance roles.
Live LinkedIn Search Apify H1B Data USCIS Database
Launch Tool →
📊
Coming Soon
Second AI product in development.
⚡
Coming Soon
Third AI product in development.
Framework Knowledge
Frameworks Applied
Every prototype in this portfolio is mapped to one or more of these frameworks. The rules aren't approximations - they're encoded in the tools.
NIST AI RMF 1.0
Applied across all four core functions. Govern maps to vendor risk. Map and Measure map to use case intake and fairness auditing. Manage maps to agent runtime governance.
GovernMapMeasureManage
EU AI Act 2024
Articles 6 and 50, Annex III, and Recital 165 encoded as deterministic classification logic in the intake tool. Risk tiers, transparency obligations, and conformity assessment rules.
Art. 6Art. 50Annex III
Singapore Agentic AI Framework · Jan 2026
All four dimensions applied to the Agent Governance Console: Risk Assessment, Human Accountability, Technical Controls, and End-User Responsibility.
Agentic AIRuntime
ISO/IEC 42001
Referenced as the AI management system standard across all prototypes for documentation, audit trails, lifecycle governance, and control recommendations.
AI ManagementAudit
EEOC 4/5ths Rule
Disparate Impact Ratio threshold (≥0.80) implemented in the Fairness Audit Tool as the primary test for adverse impact in hiring, lending, and benefits contexts.
FairnessDisparate Impact
Data Governance · Apache Atlas & Alation
Hands-on experience with enterprise data governance tooling - metadata management, data lineage, cataloging, and access policy enforcement at scale.
Apache AtlasAlation
Let's Build Something Trustworthy
Based in Seattle / Bellevue, WA. Passionate about data governance, AI risk, privacy, and security. Always interested in connecting with people building AI the right way.
LinkedIn Email GitHub