Q2B26 Copenhagen | Sam Kearney, Head of Growth, Haiqu

Q2B26 Copenhagen | Sam Kearney, Head of Growth, Haiqu

Title history
  • Q2B26 Copenhagen | Sam Kearney, Head of Growth, Haiqu — observed on 20261016-0417 UTC (PSI archive)

by Sam Kearney · View the channel on PSI

Applied Sciences & Engineering Computing & Cybersecurity UYComputer scienceUYXNatural computing
🎙 Sam Kearney 👥 6K 📅 October 12, 2026 ⏱ 19 min 👁 146 📄 expert opinion 🧭 2026-10-16
Available in: English (current) Français

Keywords

quantum computingmiddlewareCFDerror mitigationHaiqu

Summary

The talk by Sam Kearney, Head of Growth at Haiqu, presents the company’s quantum middleware software designed to make quantum computing practical for industry applications. He begins by highlighting the limitations of current classical computing and the potential of quantum devices, while acknowledging hardware challenges like noise and scaling. He draws an analogy to the rise of GPUs, crediting software (CUDA) for making them versatile. Haiqu’s approach combines an agentic layer for automating R&D workflows (e.g., literature review, algorithm selection) with a runtime layer that optimizes execution on real hardware. Key case studies include reproducing an IBM quantum utility experiment for $30 instead of $10,000, loading an Eiffel Tower image more effectively, and simulating magnetic materials with drastically reduced preprocessing time. The main focus is on computational fluid dynamics (CFD) in collaboration with Quanscience. They achieved a record largest quantum CFD simulation on real hardware (IonQ) for the Airbus-BMW challenge, then extended it to nonlinear Navier-Stokes flows with realistic geometries on IBM devices. They reduced gate counts from ~1,800 to ~540, enabling coherent results. Kearney argues this represents a step change toward quantum advantage, emphasizing that near-term devices can execute meaningful algorithms without needing hundreds of logical qubits. He closes by inviting collaboration and mentioning an academic program offering free access and AWS credits.

214 words

Critical Evaluation

Value of the Information & Strength of the Argument

The talk provides valuable insights into the current state of practical quantum computing, particularly how software layers can mitigate hardware limitations. The argumentation is structured and uses concrete, quantitative examples to support claims. However, as a company pitch, it emphasizes positive results and downplays potential caveats. The reproduction of IBM’s experiment and the CFD achievements are impressive, but the speaker acknowledges they are not quantum advantage yet. The use of analogies (GPUs) helps make complex ideas accessible, and the step-by-step progression from simple to more complex simulations builds credibility. The main weakness is that all data stems from Haiqu’s own work, lacking independent validation.

Scientific Rigor, Source Quality, Title Accuracy

The scientific rigor is moderate: the talk cites specific experiments and collaborations (e.g., IBM, IonQ, Quanscience) and mentions a Nature publication for IBM’s Ising simulation, but no peer-reviewed references for Haiqu’s own results are provided. The only external source is the Q2B conference website. The title is too generic to reflect the depth of content, though the description clarifies the topic. The speaker is open about his limited CFD expertise, which is honest but indicates the depth is not expert-level. Overall, the reliance on company-generated results and lack of external sources reduces the scientific robustness.

214 words

Title / Content Match

The generic YouTube title gives little indication of the content, but the description clearly specifies the topic. The talk itself matches the announced subject of quantum middleware for practical computing.

Quality & Reliability

7/10

The talk presents concrete case studies and quantitative improvements (e.g., cost reduction from $10,000 to $30 for IBM's Ising simulation), but these claims are not peer-reviewed and come from a company representative with a vested interest. The results are plausible but lack independent verification.

Key Moments

Markers derived by PSI from the transcript: the creator did not define chapters.

Cited Sources

  • Q2B Conference Website — The link in the description provides context for the conference and possibly further materials.

Contribution & Novelties

Haiqu’s contribution lies in offering a full-stack software solution that combines agentic automation for algorithm development with a runtime layer that minimizes resource usage on quantum devices. Their claimed results—such as drastically reducing the cost and time of running quantum experiments—have the potential to lower barriers for labs with limited budgets. The collaborative work with Quanscience demonstrates a step forward in quantum CFD by handling nonlinear, realistic flows on present-day hardware, a niche previously thought too complex.

Pour aller plus loin :

  • Quantum computing — Overview of quantum computing principles and hardware challenges.
  • Quantum error mitigation — Techniques used to improve results on noisy intermediate-scale quantum (NISQ) devices.
  • Computational fluid dynamics — Classical methods for simulating fluid flows; context for the quantum CFD efforts.
  • Quantum advantage — The threshold where quantum computers outperform classical ones; relevant to the talk’s goals.

140 words

Radar Profile

The radar profile shows strong scores on information quantity and quality, moderate technical level, and lower reliability, consistent with a company presentation that offers detailed examples but lacks independent verification. This suggests a talk valuable for understanding current quantum middleware capabilities but requiring caution regarding the objectivity of claims.

Reliability 5/10