Best Place To BuildQuantum Mechanics, qubits, superposition & superconductors with Prof. Prabha Mandayam | BP2B S2 E11
CHAPTERS
Setting the stage at IIT Madras + Prof. Prabha Mandayam’s focus
The host introduces the Best Place To Build Podcast and frames the episode as a crash course in quantum computing. Prof. Prabha Mandayam is introduced along with her work in quantum information and (especially) quantum error correction.
- •Podcast premise: meet builders at IIT Madras and learn what they’re building
- •Prof. Mandayam’s background: physicist at IITM, quantum information & error correction
- •References to her book and popular NPTEL quantum computing lectures
- •The host’s goal: understand quantum computing basics quickly and practically
From classical bits to qubits: the Bloch-sphere intuition
Prof. Mandayam contrasts classical bits (0/1) with qubits using a sphere model (Bloch sphere). The key idea is that a qubit can occupy infinitely many states between 0 and 1, which is captured by the notion of superposition.
- •Classical computing: transistors implement 0/1 and algorithms operate on bit strings
- •Qubit visualization: 0 and 1 as poles on a sphere; states span the surface
- •Two angles (like latitude/longitude) specify a qubit state
- •Superposition as the resource that expands the ‘state space’ beyond 0/1
Superposition made tangible: coin-in-flight and polarization of light
To make superposition feel less abstract, the discussion uses a coin toss analogy (capturing the coin mid-flight) and then a more physical example: photon polarization. This links the math of superposition to a real lab system people can imagine and measure.
- •Coin-in-flight analogy: quantum state as ‘not yet head or tail’ (probabilistic outcomes)
- •Biased coin illustrates amplitudes/probabilities are not always 50–50
- •Polarization example: horizontal/vertical vs 45° as a combination of both
- •Connection to vectors and decomposing a state into components (linear algebra intuition)
Why it’s called ‘quantum’: single-particle physics and photons
The word ‘quantum’ is tied to behavior at microscopic scales—single photons, single electron spins, single atoms—where outcomes become fundamentally probabilistic. The conversation explains how bulk light behaves classically, but reducing to single-photon regimes reveals quantum effects.
- •Bulk beams can be described by classical intensity; single photons cannot ‘split’
- •At single-particle scales, transmission becomes probabilistic (go through vs blocked)
- •Quantum mechanics governs these microscopic degrees of freedom
- •Quantum computing = using quantum-mechanical objects as information processors
100 years of quantum mechanics + the birth of quantum algorithms
The episode situates quantum mechanics historically (Schrödinger/Heisenberg era; roots in photoelectric effect) and then distinguishes modern quantum computing as ~40 years old. It highlights early algorithmic milestones that demonstrated computational advantage.
- •1920–1925 as ‘golden years’: Schrödinger equation, uncertainty principle
- •Earlier roots: photoelectric effect and quantization (Einstein, photons)
- •Quantum computing origin story: David Deutsch (1980s) and the Deutsch problem
- •Core motivation: quantum algorithms can outperform classical ones (speedups)
Why quantum computing mattered to the world: Shor, Grover, and RSA
The discussion moves from toy speedups to real-world impact: Shor’s factoring algorithm threatens RSA-based security, while Grover’s algorithm accelerates search. These results are presented as catalysts for major government and industry investment.
- •Shor’s algorithm: polynomial-time factoring on a quantum computer
- •Impact: RSA and public-key security rely on classical hardness of factoring
- •Complexity framing: classical vs quantum feasibility; clarifying NP vs NP-hard confusion
- •Grover’s algorithm: quadratic speedup for unstructured search; broad optimization relevance
Hardware reality check: architectures and the ‘transistor moment’
Attention shifts from algorithms to building machines—what it takes to create controllable qubits on chips. The episode surveys leading qubit architectures and explains why superconducting qubits became prominent as an engineering path toward scalable processors.
- •Need for a ‘transistor moment’ in quantum: controllable, manufacturable qubits
- •Architectures mentioned: photonics, superconducting qubits, trapped ions, neutral atoms
- •Superconducting qubit basics: superconductors at low temperature; Cooper pairs; circuit-defined 0/1 levels
- •Integrated photonics as a long-term hope for room-temperature, chip-scale systems
Decoherence: why qubits are fragile and error correction is central
Prof. Mandayam explains decoherence as the process by which quantum states leak into classical behavior due to environmental interactions. The conversation frames error correction as the defining challenge: isolate qubits enough to preserve states, but still control and measure them.
- •Decoherence = noise that collapses fragile superposition states
- •Trade-off: isolation reduces noise, but control/measurement require interaction
- •Engineering emphasis: millikelvin dilution refrigerators (‘chandelier’ setup) to suppress noise
- •Error correction as the pathway to long-lived, useful quantum computation
How many qubits are ‘enough’: from 100 today to millions for RSA-scale tasks
The host compares billions of transistors to ~100-qubit quantum chips and asks what scale is required for meaningful advantage. Prof. Mandayam clarifies that advantage is usually in time/steps, but practical factoring at scale requires large numbers of error-corrected qubits.
- •Today’s milestone: ~100-qubit proof-of-principle systems (e.g., Google)
- •Demonstrations remain small-scale (factoring 15/21) vs real cryptographic sizes
- •Rough estimates: tens of thousands of ideal qubits vs ~million with error correction overhead
- •Key message: we are early—far from large-scale fault-tolerant machines
Quantum error correction: redundancy without copying + the no-cloning theorem
Classical error correction uses redundancy by copying bits, but quantum states can’t be duplicated arbitrarily due to the no-cloning theorem. The episode explains why quantum error correction must encode information differently—protecting an entire state space, not just 0/1.
- •Classical repetition codes: protect against bit flips using majority voting
- •Quantum challenge: must protect arbitrary superpositions, not just basis states
- •No-cloning theorem: no universal ‘Xerox machine’ for unknown quantum states
- •Motivation for more sophisticated encodings than straightforward copying
Entanglement as the workaround: encoding information into the whole system
Entanglement is introduced as the key ingredient enabling quantum error correction. By spreading information across multiple qubits in a way that can’t be decomposed into independent parts, systems can detect/correct errors while respecting quantum constraints.
- •Entanglement: information resides in the joint state, not individual qubits
- •Three-qubit repetition code as a simple example with a ‘quantum twist’
- •Encoding circuit picture: one arbitrary qubit + ancilla qubits → entangled code state
- •Goal: preserve superposition and recover from noise without cloning
Will quantum computers be personal devices? Likely ‘facilities’ first
The conversation explores whether quantum computers could become handheld like smartphones. Prof. Mandayam predicts specialized shared facilities in the near term—similar to early mainframe access—before any consumer-scale transformation becomes plausible.
- •Analogy to early computing: centralized access before PCs became common
- •Near-term expectation: quantum computing centers/facilities for researchers
- •Potential application pull: optimization, simulation, protein folding
- •Uncertainty acknowledged: long-term consumer adoption is hard to predict
What to learn + programming the stack: linear algebra, VQAs, and Qiskit
The episode becomes a practical guide for learners: linear algebra is the core language of quantum computing, with probability theory as support. It also touches on variational quantum algorithms for chemistry/biology use cases and how people can program real devices via platforms like IBM Qiskit.
- •Core math: linear algebra (vector spaces, amplitudes α|0⟩+β|1⟩)
- •Probability theory as helpful secondary foundation
- •Variational quantum algorithms (VQAs) for problems like protein folding; deeper QM needed
- •Programming access: IBM’s Qiskit; full-stack abstraction still evolving across architectures
Quantum research in India: National Quantum Mission, hubs, and milestones
Prof. Mandayam outlines India’s position in the global race and argues India can still catch up, especially in hardware. She explains the National Quantum Mission structure (four hubs) and gives concrete deliverables for communication and computing targets.
- •India’s strengths: strong theory groups; growing experimental efforts
- •National Quantum Mission hubs: IIT Madras (communication), IISc (computing), IIT Bombay (sensing), IIT Delhi (materials)
- •Quantum communication goal: quantum-secure links (Chennai–Bangalore now, longer later)
- •Hardware goal examples: build and control ~50-qubit systems; fund multiple architectures to hedge bets
Prof. Mandayam’s journey into quantum + women in the field + closing advice
The episode closes with Prof. Mandayam’s personal path—from Chennai to IITM to Caltech—and her emphasis on following curiosity rather than trends. She discusses gender representation challenges and encourages students to enter quantum now as skills will remain valuable regardless of how the field evolves.
- •Personal journey: BSc in Chennai, MSc at IITM, PhD with John Preskill; early interest via Bell/entanglement
- •Choosing physics without the JEE path; supportive family and interest-driven decisions
- •Women in quantum/STEM: representation improving but still limited; societal expectations affect choices
- •Closing message: next decade is an exciting time—build math/engineering skills and join early
