// customers · production deployments

Eleven in production. Anonymized, here.

Most of our customers cannot publicly name us — yet. Mid-market doesn't issue press releases. The case studies below are real deployments, with named verticals and real numbers, identifying details removed at the customer's request.

StageLivePublic launch — free to start
Pilots converted88%cohorts 01 + 02
Avg time to first run11dayscontract → live
Net retention142%trailing 12 months
// customers · anonymized

A spread across mid-market India.

Six verticals. Eleven customers. Slot 12 onwards is reserved for cohort 03.

We had a treasury question that our analyst team had been working on for nine days. We loaded it into Godengine on a Tuesday afternoon. We had a ranked decision with a signed reasoning DAG by the coffee break. Our CFO read the trace and said two words: "It's right."
— CRO · mid-market BFSI · Tier 02 customer since Jan 2026
// case studies

Three deployments. Real numbers.

Identifying details removed. Vertical, problem shape, and quantitative outcome are accurate.

Case 01 · BFSI

Capex pacing under demand softening

A mid-sized private bank needed to phase a ₹1,200 Cr branch + tech capex program against a demand forecast that had shifted three times in two quarters.

Vertical BankingMode TitanTier 03Live since Jan 2026

The question. Should we accelerate or defer phase 2 of the capex plan, given regional demand variance and an evolving rate cycle? The CFO's team had spent six weeks building three Excel models that disagreed.

The run. A single Titan-mode invocation. 288 organs across 8 capability layers. Force-load distribution: Volatility 92, Momentum 88, Friction 81, Coupling 74, Constraint 62. Trace depth: 312 reasoning steps.

The outcome. Recommendation to defer 38% of Q4 outlay into Q1 with explicit demand-trigger gates. Signed DAG attached to the board paper. Decision approved unanimously. Six months later the gates fired in the order Godengine predicted.

Decision time9 days → 47sanalyst-week equivalent
Capex deferred (Q4→Q1)38%~₹460 Cr
Liquidity headroom held+₹620 Crin adverse regime
Case 02 · Manufacturing

Tier-2 supplier risk concentration

A tier-1 auto component manufacturer discovered, through Godengine, that 14% of their revenue was dependent on a single tier-2 node — invisible to their existing supplier scorecard.

Vertical ManufacturingMode GODTier 02Live since Mar 2026

The question. Where does our deep-tier supplier risk actually concentrate, and what is the cost of substitutability for each exposed node?

The run. GOD-mode dispatch with the Supply Chain domain pack + Coupling-force amplification. 204 organs traversed the tier graph two layers deep, modeled cascade dynamics, scored substitutability.

The outcome. A single supplier in Gujarat — known, but rated "low risk" by the conventional scorecard — was identified as the choke point for 14% of revenue. A dual-source mitigation path was scoped at +6.2% per-unit cost, payback under 7 months in expected loss reduction. Decision: dual-source, executed in 9 weeks.

Hidden concentration found14%of revenue
Mitigation premium+6.2%per-unit cost
Expected-loss reduction₹38 Cr/yrat base case
Case 03 · Insurance

Reserve adequacy under climate tail risk

A life-and-general insurer needed reserve adequacy testing for a coastal property portfolio under climate-scenario pressure. Conventional cat models did not couple to macroeconomic feedback.

Vertical InsuranceMode TitanTier 03Live since Apr 2026

The question. Are our reserves adequate for a 1-in-100 and 1-in-250 climate-coupled loss scenario, accounting for second-order economic feedback in the affected regions?

The run. Titan-mode invocation with Insurance + Climate domain packs coupled. World-Model layer (L9) ran 4,200 Monte Carlo scenarios across SSP2-4.5 and SSP3-7.0 pathways, coupled to regional economic-feedback signals.

The outcome. Reserves adequate at 0.92 confidence for 1-in-100; shortfall identified at 1-in-250 of ₹12 Cr concentrated in three coastal taluks. Mitigation: reinsurance restructure on 14% of book + diversification trigger for Tier-2 inland cities. Capital efficiency improved 8.1%.

Scenarios modeled4,200in 5m 12s wall
Reserve shortfall surfaced₹12 Cr1-in-250 scenario
Capital efficiency+8.1%post-restructure
// in their words

What they said after.

The signed trace was the unlock. I have an existing model — I just couldn't get my regulators comfortable with it. With Godengine I can hand them the DAG. The conversation gets shorter every time.
— Head of Risk · life insurer · Tier 03 customer
We thought we were buying a faster analyst. We were buying an honest one. The dissent counts on close calls have changed how my team builds consensus.
— Strategy Lead · auto component manufacturer · Tier 02 customer
Self-hosted was non-negotiable. The fact that they meant it — sealed container, no calls home, source escrow available — is why we signed without an RFP cycle.
— CIO · state government agency · Tier 04 deployment
// next cohort

Cohort 03 has three slots open.

Eight per cohort. We move slowly on purpose. Bring a real question; we'll come back inside 48 hours.

Apply for cohort 03→See pricing