Jan 2, 2019 · 25m · a16z
a16z Podcast | The Cloud Atlas to Real Quantum Computing
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this episode of the a16z Podcast, host Sonal Chokshi along with guests Jeff Cordova and Vijay Pandey explore the fundamental paradigm shifts in hardware, software, and algorithm design required to build practical quantum computing systems. They discuss how cloud accessibility, hybrid classical-quantum architectures, and hyper-exponential scaling will enable early commercial applications in computational chemistry.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. The host holds 26.4% of the talking time here. How this is scored →
The host as informed peer 5.0 Guest teaching 4.4 Guest disagreement 1.1 The host pushing back 2.9
speaking balance: gold is the host, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 3:20 Jeff corrects host on instruction languages
In an overall very agreeable conversation, Jeff offers the clearest direct reframe when Sonal asserts there is no CUDA equivalent for quantum computing, counterbalancing her with 'Well, there actually is.'
Hardest push from the host ▶ 12:56 Sonal questions premature cloud talkSonal explicitly refuses the conversational premise that quantum cloud deployment is imminent, pointing out how long classical cloud infrastructure took to reach an AWS state.
Biggest teaching moment ▶ 13:11 Jeff explains Quantum Virtual MachinesJeff educates the host on software simulators, showing that developers are already performing real quantum programming up to 30 qubits before physical hardware is widely deployed.
The host holds their own ▶ 12:06 Sonal expands on mainframe time-sharing historySonal demonstrates strong historical tech knowledge by extending the host's mainframe analogy, explaining time-sharing log sheets and how inventors fail to predict user application emergence.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | The host as informed peer | Guest teaching | Guest disagreement | The host pushing back | Why |
|---|---|---|---|---|---|---|
| Hybrid Classical-Quantum Computing and Instruction Languages | 4 | 4 | 2 | 3 | Sonal demonstrates familiarity with hardware acronyms like CPU, GPU, and TPU, and pushes on whether quantum is a discrete step rather than a continuum. Jeff gently corrects her assumption that no instruction language exists for quantum yet by introducing Quill, and reframes system stability. | |
| Probabilistic Computing and Simulating Natural Systems | 5 | 4 | 1 | 1 | Sonal connects probabilistic quantum computing to the historical evolution of statistics and natural language processing data requirements. Vijay and Jeff educate her on noise models, mixed states, and replacing deterministic trajectories with statistical inference. | |
| Hyper-Exponential Scaling and the Quantum Advantage Boundary | 4 | 5 | 1 | 1 | Vijay educates the host on hyper-exponential scaling formulas and the sudden tipping point where quantum machines surpass classical ones. Sonal frames the psychological difficulty humans have in processing non-linear, exponential acceleration. | |
| Semiconductor Manufacturing and Mainframe Cloud Analogies | 5 | 4 | 1 | 2 | Sonal probes Jeff on what specific manufacturing know-how translates to quantum hardware, then extends Vijay's mainframe analogy by discussing time-sharing and inventor unpredictability. Vijay reframes quantum hardware economics away from personal computer ownership model. | |
| Quantum Virtual Machines and Software Simulation Strategy | 6 | 5 | 1 | 5 | Sonal pushes back directly, questioning whether talking about cloud computing for quantum is premature given how long classical cloud infrastructure took to mature. Jeff educates her on Quantum Virtual Machines and software simulators that allow developers to program before physical hardware arrives. | |
| Cryogenic Physical Infrastructure and Microservice Architectures | 6 | 4 | 1 | 3 | Sonal synthesizes enterprise technology adoption behavior, highlighting how hybrid quantum architecture is a physical necessity rather than just a transition phase. Jeff explains the physical infrastructure constraints of cryogenic cooling systems that enforce cloud access. | |
| Full-Stack Software Architecture and Startup Agility | 6 | 4 | 1 | 6 | Sonal directly challenges whether building a full vertical software and hardware stack is feasible for a startup versus tech giants like Google or IBM. Jeff defends the startup advantage by emphasizing agile engineering iteration speeds over scale. | |
| Computational Chemistry Applications and Future Horizons | 4 | 5 | 1 | 2 | Vijay details algorithmic complexity scaling in computational chemistry, explaining why classical supercomputers fail at molecular simulation. Sonal prompts with target questions about classical computational limits. |
Statements from this episode (17)
Cordova: Folding@home algorithm became the primary method for protein folding
“And the interesting thing about that, at least my recollection of it, is that No one thought that algorithm would work. It didn't look anything like the previous algorithms except that it was also doing some kind of chemistry that was interesting. And then the…”
Pandey: New compute architectures require complete algorithmic rethinking
“In some ways they were right. That was impossible. It was impossible to take existing algorithms and just like shove it down to a very different architecture. You basically had to rethink the problem.”
Pandey: Folding@home was among the earliest applications running on GPUs
“And we went through this again, when GPUs came out, we actually were some of the first applications on GPUs Even before programming languages existed on GPUs.”
Cordova: Near-term quantum computers run in 100-microsecond bursts
“Unlike classical computers, which you can kind of run for days and weeks, Or in perhaps years at a time, quantum computers kind of run in bursts of a hundred microseconds.”
Cordova: Quantum processors yield different answers on every execution run
“These quantum processors actually give you different answers every time you run them.”
Cordova: Quantum computers search huge combinatorial spaces to replicate physical outcomes
“Quantum computers can actually search combinatorially a huge space for certain kinds of problems and actually find the real thing that nature would do.”
Pandey: Quantum computers will suddenly surpass classical computing at a specific threshold
“What will happen is a quantum computer at first will seem like it won't be all that useful. It will be below the number of qubits that you need. Maybe you need a hundred qubits to solve the problem and the existing machine only has 64. And so a classical compu…”
Cordova: Superconducting quantum circuits rely on standard semiconductor fabs
“The superconducting circuits we made are using standard semiconductor manufacturing technologies.”
Cordova: Cloud access will rapidly yield half a dozen quantum killer apps
“Building cloud access into how you get at a quantum computer will quicken the pace at which the killer apps are found instead of there being one, like there was in the early days of the PC and electronic spreadsheet, we might see a half a dozen of them pop up …”
Cordova: Quantum Virtual Machines Can Simulate Up To 30 Qubits
“You can run a quantum virtual machine and software up to about 30 or so qubits.”
Cordova: Quantum computers cannot store data and function purely as compute engines
“Like, you can't actually store data on these quantum computers. They're just literally, right now, just compute engines.”
Cordova: Quarter-sized quantum chips require refrigerator-sized cryogenic cooling infrastructure
“The quantum processor is like the size of a quarter. The rest of it's the size of two or three refrigerators to house it and keep it cool at like barely above zero degrees Kelvin.”
Cordova: Quantum hardware will be hosted in cloud data centers, not on-premise
“So you never really, at least in the early days, want to ever put those things, on a customer site, you want to put them in a secure facility someplace and provide cloud access or remote access to them.”
Cordova: Quantum computing's hard science is solved; the remaining challenge is engineering
“The hard science problems have been solved, but how do you build a superconducting qubit, and how do you Send it a radio frequency data to program it. The hard part now is is going through all the different ways that these things can be hooked together to buil…”
Pandey: Quantum chemistry could be quantum computing's first commercial killer app
“And I think quantum chemistry could be that application.”
Pandey: Quantum computers do not need 100,000 qubits to deliver useful chemistry simulations
“Or enzymes. It's all the chemistry, you know, so, so this is something where a quantum computer could be able to do calculations that are either much bigger at much higher accuracy where actually would be essentially much bigger at higher, much much higher acc…”
Cordova: Computer scientists do not yet know all quantum-solvable problem classes
“One of the surprising aspects of quantum computing is that scientists and mathematicians and computer engineers don't really know all of the problems that can be solved by a quantum computer. We just know some of them, and that's not How classical computing wo…”