Constraint-aware portfolio optimization, from a holdings file.

Give Gainloom a CSV of holdings and a constraint set: sector caps, single-name limits, liquidity floors, and a CVaR budget. You get back one ranked allocation with the trade-offs shown. Mean-variance and tail risk, solved in the same pass.

See a sample run
gainloom.app/optimize
1Upload holdings
2Set constraints
3Results
holdings.csv1/6 rows parsed
TickerSectorWeight
AAPLTechnology22%

Sample run · 10-name equity book · moderate mandate

Method

One pass. The objective and the constraints solved together.

Most tools make you optimize for return first, then check risk in a different screen and re-run when a limit breaks. Gainloom treats the CVaR budget and the position limits as part of the problem, not a review step.

01

Objective

Mean-variance (Markowitz) max-Sharpe, long-only. Convex and deterministic: same inputs, same book.

02

Tail risk

CVaR at 95% enters as a hard budget in the optimization, not a report you read afterward.

03

Simulation

Monte Carlo paths for drawdown and return dispersion under a chosen stress regime.

04

Constraints

Linear sector caps, single-name limits, and liquidity floors, solved in the same pass.

Positioning

Where Gainloom is meant to fit

01

BlackRock Aladdin

A risk platform you configure for months. Gainloom is an optimizer you point at a holdings file in an afternoon.

02

Bloomberg PORT

Reports the risk of a book you already built. Gainloom proposes the book, with the trade you’d place to get there.

03

Axioma

Hands you a factor risk model. Gainloom hands you the allocation and a one-line reason for every weight.

Positioning intent: Gainloom is pre-launch and not benchmarked against these platforms yet.

Worked example · Sample run · illustrative

A $250M equity book, run end to end.

A 10-name US equity universe, $250M, moderate mandate: the kind of holdings CSV you already keep in a spreadsheet.

Objective
Max Sharpe
Max sector (Tech)
35%
Max single name
18%
CVaR (95%) budget
−16%
Monte Carlo paths
12,000
gainloom.app/optimize
1Upload holdings
2Set constraints
3Results
holdings.csv1/6 rows parsed
TickerSectorWeight
AAPLTechnology22%

Sample run · 10-name equity book · moderate mandate

Security: planned at launch

What we’ll meet before any client data lands.

None of this is built yet. Gainloom is pre-launch. This is the posture we’re committing to before onboarding a single institutional client, stated now so you can hold us to it.

01

SOC 2

Type II audit planned before general availability. Not yet certified, and we won’t claim otherwise until the report is in hand.

02

Data residency

US region (AWS) at launch. EU residency available on request for European mandates.

03

Tenancy

Single-tenant isolation for institutional clients: a dedicated database per firm, no shared data plane.

04

Market & custodial data

Bloomberg and index data flows through your licensed exports; Gainloom is not a data redistributor. Custodial feeds connect through API tokens you issue and can revoke. Read-only.

05

Access & audit

Role-based access. Every optimization run is logged with its inputs, constraints, and assumptions.

FAQ

The questions a diligence team asks first.

What actually runs under the hood?+

A convex mean-variance optimizer with a CVaR (95%) tail-risk budget, linear sector and position constraints, and Monte Carlo simulation for drawdown. No black-box “AI” and nothing quantum, just the optimization stack a quant would build, wired to be usable in an afternoon.

What data can I bring in?+

The target is a holdings CSV to start (ticker, quantity, sector), then licensed Bloomberg/index exports and custodial API feeds. Nothing is ingested that you haven’t explicitly connected.

How will security and compliance work?+

Single-tenant isolation (a dedicated database per firm), role-based access, and a per-run audit log. SOC 2 Type II is planned before general availability and not yet certified. Data sits in a US AWS region at launch, with EU residency on request. Bloomberg and custodial data stay under your licenses; Gainloom reads, it does not redistribute.

Is this live yet?+

The dashboard and sample run on this page are illustrative previews, not live product output.

Who is this for?+

Allocators who find Aladdin heavy and spreadsheets fragile: RIAs, family offices, OCIOs, and small institutional books that need constraint-aware optimization without a six-month rollout.

Is this investment advice?+

No. Gainloom is decision-support software for teams that own their investment process. It proposes allocations; it does not manage money or make recommendations to end investors.

Talk to the founder directly.

Early access is going out to a small group with active allocation workflows. Join the list to shape the roadmap and get in before general availability.

No spam · no shared lists