Zibra Labs GTM brain

category

market profile

• text: Quant firms run parallel backtests and parameter sweeps where compute throughput can constrain research.
• text: Large AI workloads can span heterogeneous accelerators and cloud providers, creating scheduling and cost challenges.

personas

• title: Head of quantitative research infrastructure
  pains:

  • Backtesting capacity limits how many candidate strategies researchers can evaluate.
  • Large parameter sweeps need reliable orchestration across many nodes.
• title: Quant research lead
  pains:

  • Slow simulation turnaround delays comparison of hypotheses.
  • Compute cost limits the breadth of experiments.
• title: AI infrastructure or post-training lead
  pains:

  • Parallel rollouts and batch inference demand scalable heterogeneous compute.
  • Sourcing and managing affordable accelerators across providers adds operational work.

icp

• text: Quantitative trading firms and hedge funds running large parallel backtesting workloads; initial target, with firm size unspecified.
• text: AI labs or infrastructure teams running post-training, agent or batch inference workloads across substantial CPU and GPU fleets; broader expansion hypothesis.

message pillars

• text: Build large distributed clusters on lower-cost cloud hardware.
• text: Support quantitative simulation and parallel parameter sweeps.
• text: Extend the same compute engine to reinforcement learning.

objections

• objection: Can the runtime handle our scale and latency requirements?
  response: Benchmark an actual backtest and compare throughput, scheduling overhead and failure recovery with the current system; site figures are targets.
• objection: How do you protect proprietary strategies across clouds?
  response: Run a security and deployment review covering isolation, data residency and access controls before testing sensitive code.
• objection: We already use Ray or a managed distributed compute service.
  response: Compare total cost and throughput for a representative workload and assess migration effort rather than assuming a replacement is needed.

competitors

• name: Anyscale
  why: Managed Ray infrastructure for distributed AI compute.
• name: Coiled
  why: Managed distributed computing for parallel workloads.
• name: QuantConnect
  why: Cloud backtesting platform, an alternative for some quantitative research workflows rather than a direct runtime equivalent.

keywords

• distributed backtesting infrastructure
• quant compute cluster
• multi cloud GPU orchestration
• parallel simulation compute
• reinforcement learning infrastructure
• batch inference infrastructure

voice

• Technical and infrastructure-focused
• Ambitious about scale
• Direct and builder-oriented

signals

• text: Hiring distributed systems engineers for quant research or reporting a backtesting bottleneck suggests immediate relevance.
• text: Growing post-training or batch inference workloads across providers suggests a second potential use case.

stack

• CPU and GPU clusters
• Spot instances

hiring

• Founding Engineer, Core Runtime
• Founding ML Researcher (Post Training)

pricing

motion

channels

• Website call requests
• YC directory
• LinkedIn

confidence

Founders

• name: Ibrahim Rabbani
  title: Founder/CEO
  role: CEO
  bio: Former tech-lead of Ray. Previously building databases at LinkedIn. I've built systems that run on hundreds of thousands of machines and serve over a billion people.
• name: Zac Policzer
  title: Founder/CTO
  role: CTO
  bio: After a career making software that serves over a billion people, with a background primarily in large databases, I'm now building HPC clusters at Zibra Labs The past two years I've been focused on what the next generation of AI data center software looks like - and working to shape that future. I've been fortunate to do much of this in open source, which I'm deeply passionate about. Talk to me about databases, Mahjong, guitars, or weird horror movies. I'm currently building Zibra Labs.