Pillar product

Agentic simulations.

Populate a synthetic world with 10² to 10⁶ agents. Run it forward. Branch it. Explain it. Decide.

agentic_sim.live12,000 agents

run #81,402 · stable62 fps

01 · What it does

Three capabilities, composable.

Our runtime was built around the three things regulated institutions consistently need: reproducibility, counterfactuals, and causal attribution. Everything else follows.

01

Deterministic replay

Seed once, replay forever. Every run produces a signed manifest of inputs, agent genomes, and interaction traces. Bit-for-bit reproducible on any substrate we support.

02

Counterfactual branching

Fork a running simulation at t=k, mutate one variable, and keep both trajectories alive. Compare at the end, or at any intermediate step. Attribute outcomes to causes.

03

Heterogeneous agents

Mix LLM-backed reasoning agents with rule-based, RL, and handcrafted actors in the same population. Swap backends (OpenAI, Anthropic, local) at the agent level.

02 · Scenarios

What our customers rehearse.

Sector

Scenario

Population

Finance

Systemic contagion stress-test under Basel-IV capital regimes; counterfactual policy branches

1.2 × 10⁴

Public health

Multi-pathogen outbreak modelling with behavioural response and hospital capacity feedback

4.8 × 10⁵

Defense

Logistics resilience under contested environments; redundant routing and doctrine variants

3.1 × 10³

Energy markets

Bid-curve emergence in heterogeneous-agent power markets with renewable intermittency

2.0 × 10⁴

03 · SDK

An SDK that gets out of the way.

# define agents, environment, policy — run at scale
from bitqubic import Simulation, Agent, replay

sim = Simulation(name="macroprudential-v12", seed=0xB17Q)
sim.populate(Agent.from_manifest("cohorts/banks.yaml"), n=12_000)
sim.attach_environment("basel-iv-2026q1")

with sim.run(steps=90, parallel_branches=400) as run:
    run.fork_at(t=31, mutate={"rate_hike_bps": 75})

report = replay(run.manifest).attribute("insolvencies")
# -> signed, reproducible, regulator-ready

See a live run against your scenario.

Book a demo →