Source: https://www.quantumsteeldesign.com/proof/ Evidence before claims # Proof and automation lab. A public record of what Quantum Steel Design can demonstrate today, how steel automation will be evaluated, and where qualified human authorization remains mandatory. [Review verified evidence](https://www.quantumsteeldesign.com/proof/#verified) [Read the benchmark method](https://www.quantumsteeldesign.com/proof/#method) [Visit QSD Research](https://www.quantumsteeldesign.com/research/) Building the evidence record ## No client-result claims are being filled in with estimates. QSD does not yet have permissioned, completed client outcomes to publish. Current evidence is limited to public tools, reference assets, maintainer-run validation checks, and clearly labeled illustrative capability studies. - Verified nowPublic functions, documented data coverage, generated pages, exports, transcripts, and automated repository checks. - EvaluatingjumpSteel workflow checks and the benchmark framework on controlled, non-client material. - RecruitingDesign partners willing to establish a baseline and permit an anonymized or attributed case study. [Explore the design-partner program](https://www.quantumsteeldesign.com/design-partners/) Verified public baseline ## Evidence available for direct inspection. These are inspectable capabilities and dated validation results, not estimates of client productivity, revenue, schedule improvement, or avoided rework. Available now 2,299 ### AISC v16.0 shape reference pages The web reference exposes US customary dimensions and section properties, nominal profile previews, calculators, and downloadable formats. Maintainer checks validate all generated shape pages and their representative exports. [Open the shape reference](https://www.quantumsteeldesign.com/steel-tools/)[Inspect W18x35](https://www.quantumsteeldesign.com/steel-tools/shapes/w18x35/)[Review the AISC source ↗](https://www.aisc.org/aisc/publications/steel-construction-manual/aisc-shapes-database-v160/) Available now 20 ### Documented Shape Studio export types The offline Windows application documents 2D profiles, reports, structured data, and length-based 3D mesh outputs. It includes a command-line interface for repeatable local workflows. [Review formats, limits, and the application](https://www.quantumsteeldesign.com/steel-shape-studio/) Available now Local ### Vector PDF-to-DXF recovery PDF Rescue processes eligible vector drawing content locally in the browser. The Windows beta publishes release metadata and a SHA-256 checksum. Its page states the boundary between recoverable PDF content and original CAD authority. [Run PDF Rescue](https://www.quantumsteeldesign.com/pdf-rescue/)[Inspect release metadata](https://www.quantumsteeldesign.com/pdf-rescue-release.json) Available now Watch + read ### Steel video watch pages with written context The maintained video library pairs public YouTube videos with written context, local thumbnails, and a dedicated video sitemap. Long-form pages include transcripts and chapters. Release checks validate the current catalog and the steel-shape page embeds together. [Browse the video library](https://www.quantumsteeldesign.com/videos/)[Inspect a long-form guide](https://www.quantumsteeldesign.com/videos/how-to-read-structural-steel-drawings/) Pre-lab site audit snapshot · August 7, 2026 The repository audit reported 2,343 HTML pages, 121,833 internal links checked, 23,445 local assets checked, 2,332 structured-data blocks, and 2,373 text files checked for encoding, with zero errors and zero warnings. This is a maintainer-run technical audit of the site build, not an independent certification or a measure of detailing quality. Available now ### Connected local project records The browser workbench can save tool snapshots into projects, maintain drawing and RFI records, and export a complete project package. A useful inspection is to save two different revisions, export the project, and import it as a separate copy before relying on the backup process. [Open the dashboard](https://www.quantumsteeldesign.com/project-dashboard/)[Inspect a handoff package](https://www.quantumsteeldesign.com/project-handoff/) Available now ### PDF review without uploading source drawings Selected PDFs can be searched for embedded text, displayed side by side or overlaid, and annotated for review. Use a non-confidential pair of sample issues to inspect page selection, alignment, exported marks, and the limitations recorded on each tool page. [Search a local PDF set](https://www.quantumsteeldesign.com/drawing-search/)[Compare drawing issues](https://www.quantumsteeldesign.com/drawing-compare/) Available now ### Explicit offline downloads The offline workbench displays its included tools, release, and file size. It verifies downloaded files before replacing a previous pack. Users can check status and remove the download separately from their saved project records. [Inspect the offline controls](https://www.quantumsteeldesign.com/offline/)[Read storage and backup guidance](https://www.quantumsteeldesign.com/software/) Reusable blank benchmark record Download the header-only CSV template for recording source revisions, expected findings, baseline and assisted time, true-positive/false-positive/false-negative outcomes, unsupported cases, reviewer disposition, human signoff, and notes. It contains no results or example claims. [Download blank benchmark CSV](https://www.quantumsteeldesign.com/downloads/qsd-steel-automation-benchmark-template.csv) Published evaluation protocol ## How a claim earns its place here. Each future automation result should be tied to a frozen input set, expected outcomes, a defined scorer, recorded exceptions, and a named human approval point. Failed and unsupported cases belong in the record. 01 ### Define the task boundary State the exact input, output, intended user, supported file or data versions, prohibited uses, and consequence of a missed issue. 02 ### Freeze representative cases Version the test set and include ordinary examples, shape-family variation, incomplete inputs, ambiguous conditions, and known edge cases without exposing confidential project data. 03 ### Establish expected results Create the reference result independently of the automation where practical. Record the source, assumptions, units, reviewer, and any cases without a single defensible answer. 04 ### Run and preserve outputs Record the software and ruleset version, runtime environment, input hash, warnings, raw output, elapsed time, and any manual intervention. 05 ### Score every exception Classify correct detections, incorrect flags, misses, unsupported cases, and reviewer overrides. Do not remove difficult cases merely because they lower a headline result. 06 ### Publish context and limits Report sample size, method, metric definitions, results, failures, version, date, reviewer role, conflicts of interest, and whether the result was internal or independently reproduced. Evaluation metrics ## Evaluation metrics put accuracy first and speed second. A fast automation result is not useful if it hides uncertainty or moves an error downstream. QSD will match metrics to the task rather than select one flattering number. 01 / CATCH RATE ### Known issues found True issues detected divided by all known issues in the frozen test set, reported with the sample size and issue classes. 02 / PRECISION ### Flags that are correct Confirmed issues divided by all flags produced. Low precision creates review noise and can hide the checks that matter. 03 / MISS RATE ### Known issues missed False negatives reported by severity and category. Consequential misses are not averaged away by numerous low-risk successes. 04 / COVERAGE ### Supported cases The share of the defined task set that receives a valid result, including explicit unsupported and indeterminate outcomes. 05 / REVIEW ### Acceptance and override How often a qualified reviewer accepts, edits, rejects, or escalates the result, with the reason preserved for later analysis. 06 / CYCLE TIME ### Median and difficult cases Median and high-percentile elapsed time measured against the same baseline scope. Setup, checking, correction, and approval time remain visible. Human-accountable automation ## Five human-accountable automation gates before consequential use. Automation can assist collection, comparison, checking, and explanation. Project responsibility and release authority stay with the qualified people assigned by the contract and organization. GATE 01 ### Controlled source Confirm the governing issue, scope, units, and input provenance. GATE 02 ### Automated pass Run versioned rules and retain warnings, exceptions, and raw evidence. GATE 03 ### Evidence review Trace each proposed finding to its source rather than accepting a bare answer. GATE 04 ### Qualified decision Accept, edit, reject, or escalate based on assigned responsibility and expertise. GATE 05 ### Authorized release Release through the project's document control process and preserve the decision record. [Review the jumpSteel boundary](https://www.quantumsteeldesign.com/jumpsteel-pro/)[Read the Trust Center](https://www.quantumsteeldesign.com/trust/) Known limitations ## What this page does not prove. - QSD has no published completed-client case study, testimonial, production ROI result, or independent detailing-performance benchmark at this time. - The current site checks establish that generated assets and technical requirements are present and internally consistent; they do not certify engineering, fabrication, erection, or project suitability. - jumpSteel Pro is Quantum Steel Design's proprietary internal assisted workflow. It is not represented as an autonomous engineer, connection designer, checker of record, or fabrication-release authority. - AISC shape data and nominal profiles require competent examination for the specific application. They do not define actual mill geometry, availability, material grade, connection design, or contract requirements. - PDF Rescue recovers eligible vector paths and semantic text from a PDF. It does not recreate the original CAD model, infer missing geometry, or make a scanned image fabrication-ready. - Unless explicitly labeled otherwise, results published here are QSD-maintained and have not been independently reproduced by a third party. Open evidence register ## Future case-study positions remain visibly unfilled. These slots will be populated only after the work exists, the baseline is documented, the result is reviewed, and publication permission is clear. Recruiting ### Design partner pilot 01 Target: one bounded detailing or checking workflow with a pre-agreed baseline, success criteria, confidentiality terms, human review, and permissioned reporting. [Discuss a pilot](https://www.quantumsteeldesign.com/design-partners/) Awaiting eligible work ### First project baseline Target: compare the same defined task before and after an assisted workflow, including setup, checking, correction, approval, limitations, and exceptions. [Contact QSD](https://www.quantumsteeldesign.com/contact/) Method defined ### Controlled automation benchmark Target: publish a versioned set of non-confidential test cases with expected outputs, scoring definitions, misses, false alerts, and reproducibility notes. [Follow QSD Research](https://www.quantumsteeldesign.com/research/) Clearly labeled samples ### Current capability studies QSD's present portfolio uses fictional quantities and conditions to demonstrate intended organization and deliverables without presenting them as client history. [Review illustrative samples](https://www.quantumsteeldesign.com/portfolio/) Help create the first measured record ## Bring one bounded steel workflow. A useful pilot begins with a real problem, a controlled input, a current baseline, clear responsibility, and permission to publish only what both parties approve. [Explore a design partnership](https://www.quantumsteeldesign.com/design-partners/)