Skip to content

What if tomorrow,

Would we be ready? Or would we discover, too late, that we had prepared for the wrong future?

Bring us a problem
01

Insurance. Banking. Energy.

Three very different industries. All three depend on the same question: what might happen next?

Energy

An energy trader tests how a portfolio responds when prices swing unexpectedly.

Energy in full →
02

To answer it, they use Monte Carlo simulation.

Imagine rolling the dice, over and over again, to explore different possible futures. Each one reveals something about the risks ahead.

Fig. 2 Each die a future. Point to roll the ones nearby.
Fig. 3 To scale. The billion is ruled into millions; the thousand is the speck.
A thousand.Counted one a second: 17 minutes.
A million.Counted one a second: 12 days.
A billion.Counted one a second: 32 years.

But exploring those futures is expensive. And when asking another question costs too much, you stop asking.

03

That is why we built Henselian

Trillions of simulated paths. In seconds.

Not hours.Not days.Seconds.

Fig. 4 Futures strung from the present to the horizon. Slide the gate to compute them.
41s

1.5 trillion member-paths: a 2.5-million-member UK buy-in priced on 600,000 paths, end to end

137×

faster than hand-optimised C on all 32 threads of the same machine's CPU

28,000×*

faster than the SOA's published CPU reference on the nested variable-annuity benchmark

04

Benchmark · a UK buy-in

1.5 trillion member-paths. Every member, every future. 41 seconds.

The scheme hands over 2,500,000 members.

Every point is one member.0 of 2,500,000
1,600,000 pensioners900,000 deferreds

Each of them valued a year at a time on 600,000 futures of longevity, gilts and inflation. 100 trillion member-years. Counted one a second, the member-paths alone would take 47,500 years.

The insurer prices it.

£334.1bnabout £134,000 a member

Best-estimate liability£321.3bnPremium£334.1bnAt 1-in-200£374.7bn

The hatched band is the capital, £53.4bn: what it takes to stay standing when longevity, gilts and inflation turn together: the 1-in-200 of the 600,000 futures.

The same transfer, the same algorithm, the same machine.

Hensel, one RTX 40900:00 / 41 s
Hand-optimised C, all 32 CPU threads0:00 / 93 min

Replayed with both clocks sped up by the same 164×: the GPU bar fills in a blink; the CPU bar takes 34 seconds of your time.

Measured end to end, slowest of three runs 44.4 s. The CPU time is scaled from 640 paths on the full membership. Synthetic scheme and an illustrative mortality basis, not any scheme's figures. The 1-in-200 is the 99.5th percentile of the 600,000 path values; capital is the difference. The premium is the BEL with a 4% pricing margin.

05

It is not just about asking what if. It is about answering what then.

Fig. 5 The wall above the sea is the surplus. The dashed line is the early warning.

Solvency coverage

196%

Coverage today: 196%. The wall stands 96 points above what the regulator requires.

What if?

Rates −1 ptSpreads +1 ptInflation +1 pt

What then?

What then should we

01

What then should we protect?

The seven points of coverage still standing above the early warning.

167% against 160%

02

What then should we change?

The inflation exposure. Hedged, coverage climbs back above 180%.

the hedge, run before it is placed

03

What then should we do today?

Decide with the cost in front of you, and put the decision on record.

a quote first, then the record

Illustrative example, not a forecast.

06

On your desktop. On your servers.

Henselian is a desktop app with the engine behind it. What it does with your data, one thing at a time.

  1. 01

    Runs on macOS, Windows and Linux.

    One app on every desktop your team uses. Screens, decisions and records live on your machine.

  2. 02

    Engines on your own servers, over SSH.

    Save the servers you ssh to, then connect to one. Nothing is exposed beyond the access you already have, and a server the app has never met shows its host key first.

  3. 03

    Every decision confirmed before it goes on record.

    Approving puts a decision on the record only after the person at the machine confirms it, with Touch ID or the Mac's password. Cancel, and nothing is recorded.

    On macOS today.

  4. 04

    Keys kept in the system's secure store.

    The keys that reach your engines and your trading venue live in the Keychain, never in a config file.

    On macOS today.

  5. 05

    Files stay in two folders you can see.

    One on this machine, one on your server. Drag between them or drop files in. Nothing outside either folder can be reached.

Fig. 6 The Henselian app, drawn from its own screens, here as on macOS.

07

We are not building another dashboard that tells you what happened yesterday. We are building a simulation foundation for decisions about tomorrow.

Banks already invest in risk calculations. Insurers already depend on them. Energy traders already use them to manage uncertainty. Our direction is enterprise simulation software, bringing high-performance scenario modelling into the decision processes you already have. When computing stops being the constraint, the opportunity is not to do the same work faster. It is to ask questions that were too expensive to explore.

08

One invitation

Come and challenge us.

If you work in insurance, banking, energy, or anywhere uncertainty drives important decisions, bring us the simulation problem your organisation wishes it could solve faster. Let's find out what's possible.

Bring us a problem

The greatest danger isn't always the scenario you calculated and got wrong.

Sometimes, it's the scenario you never had time to calculate.

Fig. 7 Light before the storm. Point to turn the beam.

We cannot prevent every storm. But we can prepare before it arrives.

Henselian. Before the storm.