Skip to content

8.2 Recommended stack (Fast MVP, Scalable, Enterprise)

Stack A — Fast MVP (months 1–9, book up to ₹30 Cr)

Section titled “Stack A — Fast MVP (months 1–9, book up to ₹30 Cr)”

Goal: ship live lending in 6 – 9 months on ₹3 – 5 Cr of platform investment.

LayerChoice
LOS + LMS coreBuy from CloudBankin / Finezza / Biz2X (whichever fits cost + speed)
Underwriting rule engineUse vendor’s BRE; build only top 20 – 30 rules
Decisioning + manual reviewVendor + own light supervisor UI
KYC + KYBKarza / Signzy / IDfy (aggregator)
V-CIPIDfy / Signzy
BSAPerfios (or FinBox)
AASetu / FinBox
GSTCygnet / Webtel / Vayana
eSignLeegality / Digio
eStampSame
NACH + UPI AutoPayDigio + Razorpay
Payment gateway / payoutsRazorpay / Cashfree
SMS / WhatsAppGupshup / Kaleyra
EmailAmazon SES
DialerExotel / Knowlarity
Co-lending allocationBuild light layer on top of vendor (first partner only)
Borrower portalBuild light (React + own BFF)
Admin consoleUse vendor admin + own light additions
AnalyticsMetabase on warehouse
WarehousePostgreSQL replica + dbt (lightweight)
HostingCloud (AWS Mumbai); managed services preferred

Total platform spend year 1: ~₹3 – 6 Cr (vendor + cloud + tooling), excluding engineering team.

Stack B — Scalable (months 9–24, book ₹30 – 200 Cr)

Section titled “Stack B — Scalable (months 9–24, book ₹30 – 200 Cr)”

Goal: replace as much vendor “core” as possible with own IP; keep primitives.

LayerChoice
LOS + LMS coreBuild own progressively (start with LMS; LOS next)
Underwriting rule engineDrools or decision-tables in DB + build rule-admin tooling
DecisioningBuild
Manual reviewBuild rich case-management
KYC + KYBContinue buying (multi-vendor)
V-CIPContinue buying
BSAContinue buying (Perfios primary, FinBox secondary)
AAContinue buying (multi-AA via Setu / FinBox)
GSTContinue buying (multi-GSP)
eSign + eStampContinue buying
NACH + UPI AutoPayContinue buying
Payment + payoutsContinue buying
SMS / WhatsApp / EmailContinue buying
Dialer + collections workflowBuy dialer; build collections workflow
Field-agent appBuild
Co-lending — multi-partnerBuild allocation + settlement engine
Borrower portalBuild (full)
Admin consoleBuild (full, replace vendor)
Analytics + warehouseSnowflake / ClickHouse + dbt + Metabase
ML / model registry(start at year 2 if ready)
Workflow engineCamunda 7 / Temporal
HostingCloud, multi-AZ

Total platform spend year 2: ~₹6 – 12 Cr (vendor + cloud + tooling), excluding team.

Stack C — Enterprise (year 2+, book ₹200 – 1,000 Cr)

Section titled “Stack C — Enterprise (year 2+, book ₹200 – 1,000 Cr)”

Goal: become a platform that powers own + partner books at scale, with optional SaaS productisation.

LayerChoice
LOS + LMSOwn
Underwriting engineOwn rule engine + ML scorecards
DecisioningOwn
Co-lending — 3+ partners + DLGOwn end-to-end
KYC / KYB / V-CIPMulti-vendor (best primitives + reliability)
BSA / AA / GSTMulti-vendor
eSign / eStampMulti-vendor with intelligent routing
Mandates + repaymentsMulti-rail (NACH + eNACH + UPI AutoPay + BBPS)
CommunicationMulti-vendor
Dialer + collectionsBest-in-class
Field-agent appOwn, possibly white-label-able
Borrower / DSA / CA / partner / anchor portalsOwn
Admin console + multi-tenantOwn
Analytics + warehouseSnowflake / Databricks + dbt + Looker / Superset
ML — feature store + model registry + monitoringFeast + MLflow + own deployment
Workflow engineTemporal at scale
HostingCloud, multi-AZ + DR multi-region
ProductisationOwn, optional SaaS for external customers

What to build internally (independent of stack)

Section titled “What to build internally (independent of stack)”

These are competitive moats and should never be vendor-dependent:

  1. Policy rule library — your credit IP.
  2. Co-lending allocation + settlement — operational moat.
  3. CA / Tally distribution — channel UX that vendors don’t ship.
  4. Repeat-borrower scorecard — internal performance data is your asset.
  5. Cohort + portfolio analytics — your view, your warehouse.
  6. DLG monitoring + invocation — regulatory + operational discipline.

These are mature commodities that vendors do better:

  1. KYC primitives — PAN, Aadhaar, OVD, V-CIP.
  2. BSA parser.
  3. AA TSP integration / FIU SDK.
  4. GST APIs (require GSP licence).
  5. eSign / eStamp (require partner licence).
  6. NACH activation rail.
  7. SMS / WhatsApp / Email rails.
  8. Sponsor-bank API banking primitives (the bank itself is the rail).
TimeBuild
Month 0–3Acquisition, application, admin, integrations layer, light decisioning
Month 3–6KYC orchestration, ingestion pipeline, basic underwriting rules, sanction + docs
Month 6–9Disbursement, LMS (if not buying), collections basics, MVP analytics
Month 9–12Co-lending v1 (single partner) — allocation + settlement
Month 12–18LMS own-build (if started on vendor), enrich UW rules, fraud signals, field-agent app
Month 18–24Multi-partner co-lending, advanced analytics, ML challenger, custom warehouse
Month 24+Multi-tenancy / productisation, advanced ML, expansion products

For each module, ask:

  1. Is there a vendor that does this well? → Y/N.
  2. Is the vendor’s cost reasonable for our volume? → Y/N.
  3. Is this our competitive moat? → Y/N.
  4. Will we need deep customisation? → Y/N.
  5. Can we afford the engineering cost + risk + delay to build? → Y/N.
  6. What’s the switching cost if we buy now and build later?

The recommendation per module is the honest answer, not a default to either side.