Create 5_quantum_program_flow.md

1_concepts/5_quantum_program_flow.md

@@ -0,0 +1,36 @@
+# Quantum Program Flow
+
+There are a number of steps that generally repeat in quantum program development.  
+
+## One-time Setup Steps
+
+**IMPORTANT** For reproducibility note versions of all hardware, software, tools and libraries used.
+
+1. Define Biz Problem - is it a fit for quantum algorithms?
+2. Psuedo-code & split (quantum / classical portions)
+3. Select quantum hardware vendor - note capacity (# qubits, etc...)
+4. Select quantum language/library - note version
+5. Select cloud run env - note runtime version
+6. Setup Dev Env - local or cloud-based IDE and tools and note version (Juptyper notebook, Python, other libraries...)
+7. Review IDE visualization tools (if any) - qubit and/or register visualizers
+8. Select & configure quantum simulators - local and cloud-based
+
+## Quantum Code Development Steps
+
+1. Write quantum code, review using tools (linters, visualizers...)
+2. Run locally (debug and/or observe results with Z gates as needed)
+3. Add unit tests and re-run locally
+4. Test run on vendor quantum cloud simulator
+5. Review results from cloud simulator run for accuracy
+6. Review transpiled quantum code (which quantum gate types are used?) from cloud simulator
+7. Estimate and verify QPU & quantum gate types needed & estimated time to run
+8. Refactor code and/or reduce dataset size to reduce computational complexity
+
+## Quantum Job Run Steps
+
+1. Configure shot count for job run
+2. Submit job to queue to be run - evaluate job progress / status (can configure max allowable values, time, % QPU usage, etc...)
+3. Poll running job (on configured Cloud QPU / qubit values) to access status / completion
+4. Evaluate job run result quality
+5. Evaluate job run log files for run details (does it match your expectations for gates, number of qubits...)
+6. Evaluate job run Run overhead (compare to classical alternative algorithm)