{"id":"da6d6fe7-4fc3-46ba-9bc6-2378f5cab5eb","arxiv_id":"2501.14957","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"PyOpticL is an open-source code-to-CAD optical layout tool with dynamic beam-path routing, demonstrated on a strontium trapped-ion system that achieved 99.9% single-qubit gate fidelity.","lead":"This paper presents PyOpticL, an open-source Python library that automatically routes laser beams and places optical parts in 3D CAD layouts, and uses it to build modular 3D-printed optics plates for a strontium trapped-ion quantum computer. A generalist might read it because it shows a software-driven way to standardize and scale complex laser systems, potentially lowering the cost and complexity of quantum computing experiments.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Modular 'drop-in' claim rests on unverified design conventions; periscope failure shows conventions do not guarantee physical interchangeability.","rationale":"The paper's genuine contributions—an open-source FreeCAD-based layout tool, a working ECDL, SAS lock, and a trapped-ion laser system—are credible; the code is linked, the fidelities are plausible, and the authors acknowledge the periscope limitation. The reader's conditional verdict is fair. The most load-bearing weakness is not the fidelity error bars (which would only dent the quantitative headline) but the unsupported leap from 'one system built with these conventions' to 'modular drop-in baseplates usable for any wavelength and scalable to quantum computers.' That leap is the paper's paradigm claim. The SI design conventions are necessary for CAD-level compatibility, but compatibility of beam coordinates in a script does not guarantee physical interchangeability: manufacturing tolerances, mounting deformation, and thermal drift all enter. The periscope failure is direct evidence that the conventions do not by themselves ensure physical alignment. A single swap experiment would settle whether the conventions are sufficient. If the swap fails, the modularity claim reduces to 'custom baseplates that were aligned once,' which is not a new paradigm; if it succeeds, the conditional can be upgraded. This concern is the same one the reader identified as the weakest assumption, so the agreement is 'agree.'","tokens_in":20149,"tokens_out":6272,"duration_ms":58602,"concrete_test":"Physically build a second, nominally identical copy of one demonstrated baseplate (e.g., the single-pass AOM plate) from the same PyOpticL script and 3D print or machine it. Swap it into the working 422 nm Sr system in place of the original, without touching any other optic or realigning. Measure fiber-coupled power and single-qubit gate fidelity before and after the swap over at least 10 trials. If realignment of downstream optics is required to recover the pre-swap fidelity, or if the post-swap fidelity shifts by more than the run-to-run variance, the 'drop-in without realignment' claim fails.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central scalability claim—that PyOpticL baseplates are 'drop-in' interchangeable across subsystems, wavelengths, and scales—rests on the SI design conventions (fixed beam height, 1-inch grid, one-way branching), which are asserted to ensure that 'all baseplates are compatible and easily aligned with each other.' This sufficiency is never measured. The paper states 'We have found that in practice further automatic alignment is not necessary' (Section: Dynamic layout along beam path), but reports no swap test, no alignment tolerances, and no before/after beam-pointing or fidelity data. The only physical system built uses one set of baseplates at one wavelength (422 nm, 1/2-inch optics); the four-scale SAS demonstration (Fig. 2) is a CAD recompile, not four built systems. Moreover, the authors' own periscope exception (Rapid Prototyping section: 'they could not remain aligned for more than a day') shows the conventions do not guarantee physical stability for out-of-plane components. Thus the claim that modularity generalizes to 'any wavelength' and scales to quantum-computer-sized systems is an extrapolation from a single working example, and the central 'drop-in without realignment' benefit is exactly what remains untested.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents PyOpticL, an open-source Python library built on FreeCAD for code-to-CAD optical layout. The central idea is beam-path-based dynamic routing: optical elements are placed along beams defined by binary-indexed paths, rather than at fixed coordinates, and layouts can be recompiled with different optics dictionaries. The authors implement modular 'drop-in' baseplates for an ECDL, saturated absorption spectroscopy, and single- and double-pass AOM subsystems, and combine them into a full 422 nm laser system for a trapped 88Sr+ ion. They report Doppler cooling of the ion, 99.8% detection fidelity in 2 ms, Raman Rabi oscillations between Zeeman sublevels, and a claimed 99.9% single-qubit gate fidelity. They also show the same SAS layout compiled at four optical scales in CAD and a rendered large-scale layout inspired by a boson-sampling experiment.","tokens_in":20331,"tokens_out":5140,"duration_ms":50816,"significance":"If the technical claims hold, PyOpticL is a potentially valuable contribution to the open-source AMO and trapped-ion hardware ecosystem. The paper's main strength is an existence proof: a user can write a Python script that generates CAD models of functional optical baseplates, align them with simple conventions, and actually use the resulting laser system to cool, detect, and drive qubit transitions in a trapped ion. The four-scale CAD recompilation and the rendered large-scale layout illustrate the intended abstraction mechanism. The code is freely available, which is a genuine asset for the community. However, the quantitative support for two load-bearing numbers—the 99.8% detection fidelity and the 99.9% gate fidelity—is currently insufficient, and the central 'drop-in' modularity claim is asserted but not measured. These issues do not invalidate the tool, but they need to be addressed before the paper can support its strongest conclusions.","major_comments":[{"comment":"The claim of 'single-qubit gate fidelity of 99.9%' (Fig. 6c) is a load-bearing quantitative result, but no definition, protocol, or uncertainty is given. The paper does not state the fit function used for the Rabi oscillations, the number of experimental cycles, the Rabi pulse duration, the π-pulse calibration, or whether the quoted number is corrected for state preparation and measurement errors. As written, a reader cannot distinguish a true gate fidelity from the contrast of a sinusoid. Please either provide the complete measurement protocol and statistical error bars, or rephrase the claim as 'Rabi oscillation contrast of 99.9%' rather than 'gate fidelity'.","section":"Raman Laser for Zeeman qubit gate operations"},{"comment":"The 99.8% detection fidelity in Fig. 5b is similarly under-specified. The paper reports a single histogram without error bars, without the photon-count threshold used, and without the number of bright and dark measurements. To support the high-fidelity detection claim, please report the full bright/dark distributions, the threshold choice, and a confidence interval on the misclassification probability, or explicitly label the value as a single-run estimate.","section":"Laser cooling and detection"},{"comment":"The central 'drop-in' modularity claim is asserted but never quantitatively demonstrated. The sentence 'We have found that in practice further automatic alignment is not necessary' (Section: Code-to-CAD) is an anecdotal statement; no swap test is reported, and no beam-pointing or fiber-coupling repeatability data are given. The SI design conventions (fixed height, 1-inch grid, one-way branching) are sensible rules that make layouts easier to reason about, but they are not a measurement of interchangeability. The periscope exception in 'Rapid Prototyping with 3D Printing' is an explicit counterexample showing that the conventions do not guarantee physical stability for all components. Please either add a quantitative baseplate-swapping test (e.g., fiber-coupling power or beam position before/after replacement) or revise the abstract and main-text claims to 'can be aligned with only the two input mirrors' instead of 'drop-in without realignment.'","section":"Code-to-CAD: Dynamic Layout and Routing / SI 'Design Conventions'"},{"comment":"The four-scale SAS recompilation in Fig. 2 is a CAD exercise; only the 1/2-inch mounted version was built and tested. The abstract's claim that the baseplates 'can be used for any wavelength' is therefore an extrapolation from a single physical implementation at 422 nm. This is acceptable if clearly framed as a design capability, but the paper should explicitly state which layouts were physically realized and which were rendered only, and should temper the 'any wavelength' wording to 'designed for any wavelength within the component library' pending further demonstrations.","section":"Dynamic layout along beam path / Fig. 2"}],"minor_comments":[{"comment":"The four configurations (1-inch, 1/2-inch mounted, 1/2-inch unmounted, mini-optics) are not labeled in the figure or caption; adding a legend or text labels would improve readability.","section":"Fig. 2"},{"comment":"The Doppler scan does not show error bars or the number of repetitions; a short statement about the scan procedure and uncertainty would be helpful.","section":"Fig. 5a"},{"comment":"Please specify the beam geometry (co-propagating or counter-propagating), the measured Raman Rabi frequency, and the pulse length used for the π rotation, since these define the gate that is claimed.","section":"Raman Laser for Zeeman qubit gate operations"},{"comment":"The acknowledgments credit the MIT Quanta Lab's C4PO library as an inspiration, but the main text does not compare PyOpticL's capabilities with C4PO. A brief technical comparison would help readers understand what is new beyond the existing OpenSCAD-based approach.","section":"Introduction / Acknowledgments"},{"comment":"There is a typo in 'perisocopes' (should be 'periscopes') in the paragraph discussing 3D-printed periscope stability.","section":"Rapid Prototyping with 3D Printing"},{"comment":"The 'roughly recreate' of the boson-sampling layout in Fig. 7 is not quantitatively described; specifying the number of optical elements, the loop patterns, and any simplifying deviations would make the scalability demonstration more informative.","section":"Scalable Optical Design"}],"recommendation":"major_revision","confidential_remarks":"This is an instrumentation/tool paper rather than a conventional quantum-information theory result. The experimental demonstrations are a useful existence proof, but the quantitative claims (99.8% detection, 99.9% gate) do not yet meet the reporting standards of a quantum-information venue. The editor may want to judge whether the paper is better suited to an applied-optics or open-source-hardware journal, where the current level of metrology might be acceptable after adding error bars and clarifying the gate-fidelity definition."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: PyOpticL is a real, working library, and the authors earn the experimental claim, not just the CAD claim. The dynamic beam-path routing (binary-tree beam indexing, placement along the beam instead of fixed coordinates) is genuinely new relative to static CAD layouts like C4PO and GDSFactory. The same SAS script recompiled at four scales is a nice demonstration of the abstraction idea.\n\nWhat earns credit: they built the thing. A 422 nm ECDL with one custom 3D-printed part, an SAS lock baseplate, single- and double-pass AOM baseplates, all used to cool and detect a trapped 88Sr+ ion. The Doppler curve, 99.8% detection fidelity, and 99.9% single-qubit gate fidelity are external measurements, not outputs of the tool. The code and designs are on GitHub, which makes the work reproducible in a way most AMO hardware papers are not. Citations to prior ECDL work, C4PO, Linien, etc. are appropriate.\n\nSoft spots, in order of importance. The 'drop-in modular baseplate' claim is the load-bearing extrapolation and it is not actually tested. The SI design conventions (fixed height, 1-inch grid, one-way branching) are asserted to guarantee interchangeability, but there is no swap test, no alignment tolerance measurement, no before/after pointing or fidelity data. The four-scale demonstration is a CAD recompile, not four built systems. The periscope exception — 'they could not remain aligned for more than a day' — shows the conventions do not guarantee physical stability for out-of-plane routing. I would not call this fatal: the built system demonstrates the core value of the tool, and the authors are candid about the periscope issue. But the claim that baseplates work 'for any wavelength' and scale to quantum-computer-sized systems goes beyond the evidence.\n\nMinor issues: fidelity numbers have no error bars or shot statistics; the time-saving benefit over conventional layout is asserted, not measured. Both are fixable.\n\nThe stress-test note is on target. It does not undermine the central demonstration, just the breadth of the generalization.\n\nVerdict: this deserves a serious referee. The experimental demonstration is sufficient to justify peer review, and the modularity overclaim is the kind of thing referees can push on. The paper will be most valuable to AMO and trapped-ion groups considering open-source optical layout, and to anyone building low-cost ECDLs. I'd bring it to reading group and would cite it if I were building a laser system. Send to review.","headline":"PyOpticL is a real open-source tool with a working trapped-ion demo behind it; the drop-in modularity claim outruns the evidence but not by much.","tokens_in":20905,"tokens_out":2115,"would_cite":true,"duration_ms":19916,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A Python library called PyOpticL routes laser beams dynamically in CAD, letting optical subsystems be compiled as modular 'drop-in' baseplates; the authors use it to build a strontium trapped-ion laser system that achieves 99.8% detection…","keywords":["optical layout","code-to-CAD","modular baseplates","trapped ions","laser systems","quantum computing hardware","open-source hardware","beam-path routing"],"falsifier":"Swap two independently designed baseplates built under the stated design rules into the strontium setup — e.g., replace the rubidium lock plate with another compiled at a different optics scale or wavelength — and measure detection and gate fidelity without any realignment step; if the ion is lost or fidelity drops measurably below the reported 99.8% and 99.9%, the claimed drop-in interchangeability fails.","tokens_in":86,"feed_emoji":"⚛️","tokens_out":9414,"duration_ms":151883,"temperature":0.7,"pith_summary":"This paper claims that optical systems can be engineered like electronic circuits: instead of placing and aligning every component on a table by hand, a Python library called PyOpticL traces laser beams through the system and places components along the beam path automatically, so a layout script compiles into a physical CAD model. The authors argue this dynamic beam routing removes the bottleneck that has kept atomic and molecular optics experiments from being modularized and scaled up. To support the claim, they design modular 'drop-in' baseplates for common laser subsystems and assemble them into a complete 422 nm laser system for trapped strontium ions, reporting 99.8% state-detection fidelity and 99.9% single-qubit gate fidelity. If the approach generalizes, quantum-computing optics could be shared, version-controlled, and recompiled for different wavelengths and sizes the way microchip layouts are.","feed_headline":"99.9% gate fidelity from a code-to-CAD laser system","feed_subtitle":"Open-source PyOpticL routes beams automatically, so the same baseplate design recompiles for any wavelength or optics size.","key_machinery":"The central mechanism is beam-based routing with a binary-tree beam index: each beam segment carries a binary number (starting at 0b1; 0 for transmitted and 1 for reflected at each splitter), which gives every sub-beam a unique, readable identifier. Components are then placed relative to a beam segment by distance or cardinal direction, with no fixed coordinates, so when any element moves the whole routing re-simulates and the layout stays connected. A second load-bearing piece is a set of design conventions: all beams at one height, all beams on a 1-inch grid, and one-way branching, which together let baseplates from different scripts be joined as drop-in modules. The library is built on an open-source CAD engine with a Python back-end, which is what makes runtime variable updates and parametric recompilation possible.","core_discovery":"The paper's central claim is that dynamic, beam-based routing makes optical layout scriptable and modular: PyOpticL represents each laser beam as a path carrying a unique binary index — at each splitter the transmitted beam appends 0 and the reflected beam appends 1 — and optical elements are placed along a beam segment rather than at absolute coordinates. Because the layout is defined by code and re-rendered by the underlying CAD engine, the same saturated-absorption-spectroscopy layout script can be recompiled with 1-inch, 1/2-inch mounted, mount-free 1/2-inch, or 3-mm micro-optics by swapping a dictionary of components. The authors build three kinds of modular baseplates (a diode-laser source, acousto-optic intensity and frequency modulators, and a rubidium saturated-absorption lock), combine them into a dynamic full laser system, and use it to laser-cool a single trapped strontium ion, detect its state with 99.8% fidelity, and drive Zeeman-qubit Raman transitions with 99.9% single-qubit gate fidelity. The point is not just that these particular plates work, but that the code-to-CAD architecture makes the optics recompilable and abstractable, so subsystems can be treated as black boxes in larger experiments.","pith_inferences":["If the design conventions (fixed height, 1-inch grid, one-way branching) generalize, the library's value compounds with library size: every new baseplate becomes compatible with every existing one, so the hard part shifts from alignment to component metadata and beam-path logic.","The paper only demonstrates planar, low-profile plates; the stated limitation that 3D-printed periscopes cannot stay aligned suggests that whether the drop-in claim extends to 3D-routed layouts is an open, testable question.","A testable extension would apply the same code-to-CAD approach to other domains where free-space layouts must be reused across wavelengths, such as compact atomic clocks or photonic packaging, where the bottleneck is similarly layout reconfiguration rather than individual component design.","One implicit consequence is that alignment skill is transferred from the laboratory bench into the component library: once a mount's optical center is defined in code, anyone can compile the same layout without hands-on optical training, which could lower the barrier to entry for AMO experiments."],"forward_implications":["The same baseplate designs can be recompiled at any wavelength and with different optics sizes, so a laser-cooling subsystem for strontium could be adapted to rubidium or neutral-atom experiments without redesign.","Optical subsystems become black boxes with standardized inputs and outputs, enabling hierarchical abstraction like VLSI design; the paper's example recreates a large interferometry layout using for-loops.","3D-printed baseplates let researchers prototype layouts in resin and then machine the same plates in aluminum, reducing iteration time and cost; the paper reports a diode laser at roughly one-fifth the cost of commercial lasers.","Because layouts are scripts, hardware designs can be managed with version control and shared as open-source modules, creating a collaborative ecosystem similar to what exists for photonics layout.","The demonstrated detection and gate fidelities (99.8% and 99.9%) show the modular system performs at a level adequate for precision AMO and ion-trap quantum-computing applications."],"supporting_citations":[{"why":"Supplies the parametric CAD engine and Python scripting back-end on which the library is built.","marker":"[8]"},{"why":"Precedent for code-driven, open-source layout tools in photonics that this library extends to free-space optics.","marker":"[6]"},{"why":"Establishes the open-source control-hardware ecosystem the work aims to complement and contribute to.","marker":"[9]"},{"why":"The open-source locking software used to stabilize the laser to rubidium in the demonstration.","marker":"[36]"},{"why":"Provide the strontium-ion level structure used for Doppler cooling and fluorescence detection.","marker":"[34, 35]"},{"why":"Framework for Zeeman qubit gates and magnetic-field-noise-limited coherence used in the Raman experiments.","marker":"[38]"},{"why":"Spontaneous-emission error model justifying the large Raman detuning for high-fidelity gates.","marker":"[39]"},{"why":"The large interferometry layout recreated with for-loops to demonstrate scaling of the design approach.","marker":"[43]"}],"fun_headline_variants":["PyOpticL: code-to-CAD laser system hits 99.9% gate fidelity","Recompile optics for any size: PyOpticL's modular plates achieve 99.9% gates","From code to CAD: PyOpticL makes optics recompilable, hits 99.9% gate fidelity","Open-source PyOpticL turns laser design into code, with 99.9% qubit gates","PyOpticL: scriptable optics, 99.9% gate fidelity"],"cache_read_input_tokens":23040,"weakest_assumption_plain":"The whole 'drop-in baseplate' story rests on the assumption that the design rules — all beams at one height, on a 1-inch grid, branching one way — make any baseplate built under them align with any other without rework, a property demonstrated for the specific planar plates in this paper but not for other wavelengths, part sizes, or 3D-routed layouts.","fun_headline_variants_meta":{"raw":{"variants":["PyOpticL: code-to-CAD laser system hits 99.9% gate fidelity","Recompile optics for any size: PyOpticL's modular plates achieve 99.9% gates","From code to CAD: PyOpticL makes optics recompilable, hits 99.9% gate fidelity","Open-source PyOpticL turns laser design into code, with 99.9% qubit gates","PyOpticL: scriptable optics, 99.9% gate fidelity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00067,"raw_usage":{"total_tokens":3126,"prompt_tokens":1090,"completion_tokens":2036,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":706,"completion_tokens_details":{"reasoning_tokens":1909}},"tokens_in":706,"tokens_out":2036,"duration_ms":14027,"temperature":1.0,"reasoning_tokens":1909,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T14:44:55.628176+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Swap two independently designed baseplates built under the stated design rules into the strontium setup — e.g., replace the rubidium lock plate with another compiled at a different optics scale or wavelength — and measure detection and gate fidelity without any realignment step; if the ion is lost or fidelity drops measurably below the reported 99.8% and 99.9%, the claimed drop-in interchangeability fails.","supporting_citations":[{"cited_title":"Maurice, Z","cited_arxiv_id":null,"evidence_quote":"Supplies the parametric CAD engine and Python scripting back-end on which the library is built."},{"cited_title":"Zhang, J","cited_arxiv_id":null,"evidence_quote":"Precedent for code-driven, open-source layout tools in photonics that this library extends to free-space optics."},{"cited_title":"Strangfeld, S","cited_arxiv_id":null,"evidence_quote":"Establishes the open-source control-hardware ecosystem the work aims to complement and contribute to."},{"cited_title":"Daffurn, R","cited_arxiv_id":null,"evidence_quote":"The open-source locking software used to stabilize the laser to rubidium in the demonstration."},{"cited_title":"Akerman, Trapped ions and free photons, Ph.D","cited_arxiv_id":null,"evidence_quote":"Framework for Zeeman qubit gates and magnetic-field-noise-limited coherence used in the Raman experiments."},{"cited_title":"Madej, L","cited_arxiv_id":null,"evidence_quote":"Spontaneous-emission error model justifying the large Raman detuning for high-fidelity gates."},{"cited_title":"Ozeri, W","cited_arxiv_id":null,"evidence_quote":"The large interferometry layout recreated with for-loops to demonstrate scaling of the design approach."}],"review_version":1}