Explainable margin-risk OS for Indian brokers launching leveraged products without building a full real-time risk stack.

1. Retail trading volume is already large enough that brokers can justify dedicated risk infrastructure instead of manual ops.
2. Fresh funding is being deployed into margin trade funding and AI features, creating immediate product-roadmap demand for safer decisioning layers.
3. Competition with Dhan, Groww, and Zerodha increases pressure to ship differentiated active-trader products faster without operational mistakes.
4. Active-investor platforms are competing on execution workflows and instant order experiences, which elevates the value of low-latency backend decision engines.

Catalyst. Sahi’s funding, scale, and stated push into margin trade funding and AI features suggest brokers now have both the demand and urgency to modernize risk decisioning instead of treating it as back-office plumbing.

The product plugs into a broker’s order stream, positions ledger, collateral data, and customer profile systems to calculate account-level leverage eligibility in real time. It gives risk and product teams configurable policies for exposure, concentration, cooling-off rules, and segment-specific prompts without forcing a full internal rebuild. Every approval, block, or customer-facing nudge is stored with an explainable rationale and policy version so teams can review incidents and tune rules faster. The initial product avoids generating open-ended trade advice; it focuses on policy-bound eligibility and workflow recommendations that are easier to govern. Over time, the same control plane can power safer cross-sell and retention journeys across active-investor products.

What's different. This is not another retail trading app or generic AI copilot. It is broker infrastructure built around the exact moment when a firm expands from low-friction broking into leveraged and AI-assisted products, where explainability and policy control matter more than flashy UX. The defensible layer comes from broker-specific policy templates, integration depth into trading and collateral systems, and the feedback data generated from real approval, block, and prompt outcomes.

Jobs to be done

flowchart LR
  Buyer[Broker risk and product team] --> Pain[Manual margin approvals and weak guardrails]
  Pain --> Product[Explainable MTF risk engine]
  Product --> Outcome[Faster launches and safer active trader growth]

Executive takeaways

• Indian retail broking is already large enough to justify specialized infrastructure: CDSL reported 15.29+ crore investor accounts and 3.73 crore new demat accounts opened in FY25, while NSE active clients rose 21% in FY25. [2][3]
• MTF is no longer a niche feature. HDFC Securities said industry-wide MTF balances rose from about ₹50,000 crore in March 2024 to a peak of ₹85,000 crore in September 2024 before stabilizing near ₹75,000 crore by February 2025. [11]
• Broker economics support software spend: major brokers publicly market 4x leverage and double-digit annualized MTF interest rates, and Angel One disclosed client funding as 11.6% of FY25 revenue mix. [12][14][15][16][23][24][25]
• Regulation raises the value of an explainable overlay rather than killing the category: SEBI has tightened pledge/re-pledge, MTF broker certification, retail algo participation, and cyber-resilience requirements. [4][5][7][8][9]
• The obvious alternatives are full-stack OMS/RMS vendors or internal builds, but both skew toward broader trading infrastructure rather than a fast-to-deploy, broker-specific, auditable MTF decision layer. [17][18][19][20]
• The biggest headwind is not demand but risk appetite and compliance optics: SEBI said 93% of individual traders lost money in equity F&O between FY22 and FY24, so any product that looks like advice or weak leverage control will face scrutiny. [6]
• Sahi's new funding and stated expansion into margin trade funding and AI features is a current validation signal that Indian brokers are actively adding exactly the workflows this startup targets. [1][21][22]

Market definition

This market is broker-facing decision infrastructure for SEBI-regulated Indian brokers and active-investor platforms launching or expanding leveraged products, starting with Margin Trading Facility eligibility, collateral monitoring, concentration controls, and auditable next-best-action prompts. It excludes consumer brokerage apps, exchange core risk systems, generic CRM tools, and open-ended trade-advice products.

Customer and buyer

Primary ICP: SEBI-registered Indian brokers with enough active-trader volume to justify MTF but not the engineering depth of the largest incumbents. Economic buyer is typically the CRO, COO, or VP Product; day-to-day users are risk operations, compliance, treasury/client-funding, and product teams responsible for MTF rollouts and incident review.

Buying triggers

• A broker is preparing its first serious MTF rollout or widening funded-stock coverage and must meet exchange/SEBI approval, pledge, and monitoring requirements without building a full stack internally. [8][9][7]
• A volatility spike or loss event forces risk/compliance teams to explain approval, rejection, or restriction decisions faster than spreadsheet-based reviews allow. [6][11]
• Competitive pressure from fast-growing retail brokers pushes product teams to add leverage and AI-assisted workflows on tighter timelines. [1][2][21]

Willingness to pay

Buyer budget should exist because MTF is already monetized as a spread product: public broker materials show annualized MTF rates from about 9.65% to 15.72%+ and leverage up to 4x-5x, while Angel One and HDFC Securities both describe client funding / MTF as meaningful growth areas. That makes a product-, risk-, or client-funding-funded software budget plausible even before broader expansion modules. [11][12][14][15][16][23][24][25]

Category dynamics

Growth signal 21% FY25 growth in NSE active clients

Validation signals

• Sahi raised $33M and explicitly plans to expand into margin trade funding and AI-led features.
• CDSL crossed 15.29+ crore investor accounts and added about 3.73 crore demat accounts in FY25.
• NSE active clients grew 21% in FY25, with Groww, Zerodha, and Angel One at multi-million active-client scale.
• HDFC Securities reported its own MTF portfolio up 50% YoY in FY25 and described industry-wide MTF balances peaking at about ₹85,000 crore.
• Angel One highlighted a dedicated MTF section and a growing client funding book in FY25.

Regulatory & technical constraints

• MTF collateral must work within SEBI's pledge / re-pledge framework, so the product has to integrate cleanly with depository-linked collateral states rather than invent parallel records.
• Brokers offering MTF face exchange approval and net-worth certification requirements, increasing compliance involvement in any rollout.
• SEBI's retail algorithmic trading safeguards make it risky for the startup to drift into open-ended recommendation behavior.
• CSCRF raises baseline expectations for logging, access control, resilience, and incident processes for vendor software touching regulated entities.
• Latency and reliability matter because the underlying broker experience is increasingly sold on execution speed and instant decisioning.

The direct market is not crowded with India-native point solutions for explainable MTF decisioning, but the substitute set is strong: incumbent OMS/RMS vendors, global brokerage-platform providers, and internal rule engines. That means the startup must sell as an overlay that lands faster and with more policy transparency than a full-stack platform swap.

Why incumbents do not win by default

• Legacy OMS/RMS incumbents. Vendors like Symphony and ODIN win when a broker wants broad trading infrastructure, but their scope is much wider than a focused, explainable MTF policy layer; that creates room for a faster deployment and better audit UX.
• Global broker-platform vendors. Devexperts and similar platforms bring strong engineering and risk tooling, but India-specific SEBI, exchange, and depository workflows still create localization and implementation work a specialist can package better.
• Workflow / CRM suites. FYNXT-style platforms can automate onboarding and broker operations, yet they are not narrowly optimized for intraday collateral health, concentration limits, and Indian MTF explainability.
• In-house builds. Top brokers can build internally, but mid-market brokers still face data-fragmentation, cyber-compliance, and audit requirements that make a pre-integrated control plane attractive if time-to-launch is materially shorter.

This company sells an explainable margin-risk decision layer to SEBI-registered Indian brokers launching or expanding Margin Trading Facility and adjacent leveraged workflows. The immediate customer is a mid-market broker with roughly 100,000-300,000 demat accounts whose product and risk teams are trying to launch MTF without rebuilding a full real-time risk stack. The first product is a read-mostly overlay that ingests order, position, collateral, and customer-profile data to score MTF eligibility, monitor concentration and collateral health, and log every approval, block, and pre-approved nudge with policy versioning. The go-to-market system is founder-led sales into CRO / COO / VP Product, triggered by an MTF rollout, a volatility incident, or competitive pressure to match peers shipping leverage features faster. Research supports demand signals: Indian retail trading scale is large, industry MTF balances reached roughly ₹75,000 crore by February 2025, and brokers openly monetize leverage through double-digit MTF rates. The plan deliberately avoids consumer trading UX, open-ended trade advice, and full OMS replacement; the wedge is a faster-to-deploy overlay that is easier to audit than in-house rules plus spreadsheets. The strongest downside risks are a small near-term beachhead, regulatory sensitivity if prompts look like advice, and the possibility that brokers prefer incumbent platform renewals or internal builds. Based on current research, the near-term market is real but narrow (estimated TAM $14.5M, SAM $6.5M, SOM $1.4M), so the investable question is whether MTF control can expand into broader broker decision infrastructure fast enough.

Problem

• Mid-market Indian brokers need MTF and adjacent leveraged products to compete, but many still rely on fragmented data, spreadsheet exception handling, and in-house rule tables.
• Those workflows make approvals slow during volatility, obscure why a customer was approved or blocked, and increase credit, concentration, compliance, and incident-review risk.
• Full OMS/RMS replacements are heavy projects, while current substitutes are too broad or too manual for a fast MTF launch.

Solution

• A broker overlay that connects to order, position, collateral, and customer-profile systems to calculate real-time MTF eligibility, collateral health, and concentration limits.
• Operator console with configurable policies, case review, immutable decision logs, and policy versioning so every approval, block, or pre-approved prompt is audit-ready.
• Initial scope stays inside infrastructure: policy-bound eligibility and workflow nudges, not open-ended investment advice or retail trading UX.

Why we win

• The product lands as an overlay, so a broker can modernize risk decisioning without replacing its OMS/RMS stack.
• Explainability and policy control are core product features, which matches SEBI-driven audit and conduct requirements better than generic rules engines or AI copilots.
• Broker-specific policy templates, reusable connectors, and decision-outcome history create switching costs once the system is live.

Milestones

flowchart LR
  Wedge[Audit-ready MTF overlay] --> MVP[Shadow-mode eligibility and audit trail]
  MVP --> Proof[Paid pilots and faster broker launches]
  Proof --> Expansion[Automated controls and adjacent broker workflows]

Founding team

Experiment roadmap

Risk assessment

What must be true

• At least 3 of the first 10 target brokers say MTF approval or collateral monitoring is a top-3 launch blocker.
• At least 2 brokers prefer an overlay to an OMS replacement if deployment is under 8 weeks.
• The MVP can cover more than 80% of MTF approval decisions in shadow mode with read-only feeds and manual fallback.
• External compliance review confirms policy-bound prompts and audit logs stay outside the investment-advice boundary.
• At least 1 pilot converts to production at or above ₹60 lakh annualized software spend within 6 months of go-live.

Open diligence questions

• Which OMS, back-office, and depository combinations cover the first 10 target logos?
• Who owns the MTF P&L and signs the software budget inside a target broker?
• What percentage of current MTF approvals and exception reviews are still manual at target brokers?
• How long do incumbent OMS/RMS change requests take compared with an overlay deployment?
• Which prompt types will customer compliance teams approve, restrict, or ban?

Model sanity

• Revenue engine. The base case is driven by two paid pilots in Y1, 2-3 production brokers by Q4Y2, and 4.8 production-equivalent customers by Q4Y3.
• Must go right. Pilot-to-production conversion has to stay at or above the business plan's 50%+ target so the company reaches repeatable broker deployments before cash buffer becomes strategic.
• Model breaks if. A 9-month sales cycle or a 15% ARPU cut removes roughly $160-180K of Y3 revenue and pushes cash toward the downside-case low point.
• Next-round proof. The next financing is justified only after 2-3 production logos, one channel partner, and at least one adjacent-module sale prove the company can expand beyond a narrow MTF wedge.

Scenarios

Sensitivity

flowchart LR
  TargetAccounts --> PaidPilots
  PaidPilots --> ProductionLogos
  ProductionLogos --> ExpansionModules
  ProductionLogos --> Revenue
  ExpansionModules --> Revenue
  Revenue --> GrossProfit
  GrossProfit --> Cash

Flags: Revenue is still concentrated in fewer than 5 broker relationships by the end of Y3. · Revenue per FTE is below typical SaaS benchmarks, so the model needs either higher ACV or more productized delivery. · The beachhead market is narrow, so adjacent-module pull by Y2 is required for a venture-scale narrative. · Base case remains EBITDA-negative through Y3, so a delayed pilot conversion can pull the next raise forward.

• Regulatory boundary risk. The product could be viewed as crossing from risk tooling into investment advice if prompts become too prescriptive. Mitigation: Start with policy-bound eligibility and guardrail workflows, keep recommendation logic configurable by the broker, and maintain explicit audit logs.
• Incumbent internal build. Larger brokers may prefer to build their own risk stack once the need is obvious. Mitigation: Target mid-market brokers first, minimize deployment time, and build reusable integrations and policy templates that are hard to match quickly.
• Integration drag. Broker data is often fragmented across order, ledger, collateral, and compliance systems, slowing time to value. Mitigation: Begin with read-only integrations for the highest-value MTF decisions, support a narrow vendor set first, and package implementation services tightly.