Pennant Technologies Founding Platform Partner

Hyderabad, September 2026 — Pennant Technologies Private Limited, one of India's most trusted BFSI software companies, has integrated Zytra's AI governance platform into the pennApps Lending Factory as a founding platform partner. The integration brings continuous AI risk governance, runtime guardrails, and RBI-compliant evidence trails to every bank, housing finance company, and NBFC running on the Pennant stack.

14M+
Loans under governance on the Pennant platform
100M+
Annual transactions with AI risk coverage
4 of top-10
India HFCs running pennApps
4 of top-10
India NBFCs running pennApps

Why This Partnership

The AI governance gap in Indian banking is regulatory, not theoretical. RBI's AI/ML Circular (January 2025) and the draft eMRM framework (June 2026) require financial institutions to maintain continuous AI risk inventories, produce audit-trail evidence on demand, and demonstrate that AI systems operating in regulated workflows have undergone independent governance review — not just model validation.

For banks running on Pennant's platform, that governance obligation now extends to every AI-assisted decision within the lending workflow: credit scoring models, document extraction pipelines, fraud detection systems, collections agents. Governance is no longer optional for these deployments — it is a regulatory precondition for continued operation.

Pennant's partnership with Zytra closes that gap at the platform level. Rather than each bank independently deploying a governance layer on top of Pennant's stack, the integration makes AI governance a native capability of pennApps — available to every Pennant customer without a separate implementation project.

"Banks are asking us for governance capability before they can expand AI use in their lending workflows. They're not asking their AI vendors — they're asking us, as the platform that runs the workflow. Our partnership with Zytra means we can give them that answer, with a solution that was built for BFSI from day one, not adapted from a generic enterprise governance tool."

— Rama Krishna Raju, Founding Director & CEO, Pennant Technologies

How the Integration Works

The Pennant–Zytra integration operates at three layers within the pennApps Lending Factory:

Layer 1

AI Asset Registry — continuous inventory

Every AI model and AI-assisted workflow within pennApps is automatically registered in Zytra's AI Asset Registry. Credit scoring models, document extraction pipelines, fraud detection systems, and customer-service agents are inventoried with their risk tier, validation status, and regulatory mapping — updated continuously, not point-in-time.

Layer 2

Aval Runtime Guardrails — inference-time protection

Zytra's Aval guardrail layer sits inline with pennApps AI inference paths, applying real-time classification across BFSI threat categories — prompt injection, PII exfiltration, financial advice violations, regulatory breach — at 11.6ms latency. For collections agents and other agentic deployments within pennApps, Prism governs multi-turn and multi-step workflows with authorization boundaries and audit logging.

Layer 3

Evidence Trail — RBI-ready audit packs

Every governance action — risk assessment, control assignment, validation sign-off, guardrail event — is written to a hash-chained evidence ledger. Banks can generate RBI examination-ready audit packs on demand: PDF, OSCAL, and regulator-ready evidence exports covering the full AI risk lifecycle, without manual coordination before an inspection.

What This Means for Pennant Customers

Banks and NBFCs on pennApps now get

  • Continuous AI risk inventory — no manual registry maintenance before inspections
  • RBI eMRM-aligned evidence packs — clause-mapped to RBI AI/ML Circular (Jan 2025) and SEBI Algo-AI Guidelines
  • Runtime guardrails on lending AI — prompt injection, PII, financial advice enforcement at the inference layer
  • Agentic workflow governance — authorization boundaries and audit trails for collections agents and multi-step document processing
  • Board-level residual risk reporting — heat maps by risk domain, not just model-level validation status

For the CROs and model risk officers at Pennant's bank customers, the practical impact is concrete: the AI risk register that previously required weeks of coordination before an RBI inspection is now maintained automatically. Evidence packs are generated in minutes, not assembled under deadline pressure. The governance infrastructure that regulators require exists continuously — not as a periodic exercise.

"Every bank we speak to in India is trying to figure out how to govern AI without building a separate governance infrastructure from scratch. Pennant is already running their most critical lending workflows. Integrating Zytra at the platform level means their customers get enterprise AI governance without a separate deployment project. That's the right architecture for India BFSI."

— Zytra Research Team, Zytra Technologies

About Pennant Technologies

Pennant Technologies Private Limited builds mission-critical software for banks and financial institutions. The company's flagship product, pennApps Lending Factory, powers lending operations — from loan origination and credit decisioning through collections, servicing, and regulatory reporting — for banks, housing finance companies, and NBFCs across India and Asia. Four of India's top-10 HFCs and four of the top-10 NBFCs run on the Pennant portfolio. The company was founded by Rama Krishna Raju, who brings 25+ years of building world-class BFSI products.

pennant.net

About Zytra Technologies

Zytra builds AI governance and safety for banking and financial services. Zytra's platform — Aval (runtime guardrails), Prism (agentic governance), and the AI Governance Module — is purpose-built for BFSI: clause-mapped to RBI, SEBI, DPDP Act, EU AI Act, NIST AI RMF, and ISO 42001, and deployable in the bank's own tenant. The open FinProof benchmark independently measures BFSI AI safety performance. Aval v1.5 holds the #1 position: 0.5% AgentHarm false-positive rate at 11.6ms inference latency.

For the full case study on the Pennant partnership, see: Pennant Technologies × Zytra: AI Governance at Scale →

Media contact: For interview requests or additional information, use the contact form. For partnership enquiries from BFSI software vendors, see Zytra Partners.