ERWIQ — The Intelligent Engineer

Design aircraft from mission to report.

Building the AI-native operating system for aerospace engineering — established methods, orchestrated by AI, every result traceable to the tool run that produced it.

  1. Mission
  2. Design
  3. Analyze
  4. Report

The problem

Conceptual design lives in a dozen disconnected tools.

Requirements in spreadsheets, sizing scripts in MATLAB and Python, aero runs in XFOIL and AVL, and the results stitched into documents and slide decks by hand. Every handoff is manual — data re-entered, assumptions lost along the way.

  1. Excel
  2. MATLAB
  3. Python
  4. XFOIL
  5. AVL
  6. Reports
  7. PowerPoint
  1. Mission
  2. Engineering Engine
  3. Aircraft Design
  4. Report

The same established methods, run as one orchestrated flow — with every result traceable to the tool run that produced it.

Platform

One platform for the conceptual design loop.

From mission parsing to the engineering report — six capabilities that run as one flow.

  • Mission Intelligence

    Describe the mission in plain language — range, payload, endurance, launch constraints. ERWIQ parses it into a structured design spec.

  • Aircraft Design

    Conceptual sizing runs as an orchestrated sequence — wing sizing, weight buildup, geometry — not improvised numbers.

  • Performance Analysis

    Aerodynamic and stability estimates with interactive charts for every design run, refined across iterations in the same session.

  • Engineering Reports

    A structured PDF engineering report generated on demand from actual run results — requirement, sizing, aero, propulsion, geometry.

  • Engineering Trace

    Every number links back to the tool run that produced it, and refinements carry values forward with explicit provenance badges.

  • Component Selection

    Airfoil selection plus motor, propeller, battery, and ESC sizing matched to the mission's power and endurance requirements.

How it works

Requirement to report.

One design run, every step explicit — the same numbers you see are the numbers the tools produced.

  1. Mission

    mission parsing

    Describe the requirement in plain language — range, payload, endurance, launch constraints.

  2. Requirements

    design spec

    The mission is parsed into a structured design spec — the requirements every downstream tool works from.

  3. Sizing

    size_aircraft

    Initial aircraft sizing from the mission: wing area, span, aspect ratio, wing loading, cruise and stall points.

  4. Weight

    size_aircraft

    A mass buildup — gross, empty, battery — kept consistent as the design iterates.

  5. Battery

    size_propulsion

    Battery energy and capacity sized for the endurance target with reserve margin — chemistry, voltage, and mass included.

  6. Propulsion

    size_propulsion

    Motor class, propeller, and ESC matched to the power the mission actually demands.

  7. Performance

    select_airfoil · run_aero_analysis

    Airfoil selection and aerodynamic analysis — polars, L/D, trim — with interactive charts for every run.

  8. Geometry

    generate_geometry

    An outer mold line generated from the sized dimensions, viewable in 3D.

  9. Report

    report generator

    A structured PDF engineering report, generated on demand from the actual run results.

The product

Numbers with provenance.

A real design session in ERWIQ — UAV conceptual sizing with charts, and every section labeled as recomputed or carried forward when a design is refined.

ERWIQ results view: sizing stat tiles, drag polar and lift-to-drag charts, with provenance badges marking sections as updated this iteration or carried from iteration 1
Actual output from a completed design session, unedited.

Under the hood

Real tools, orchestrated.

The engineering is built on established, citable methods — the AI's job is to run them in the right sequence and explain the results.

  • AVLMIT · Drela

    Vortex-lattice aerodynamic & stability analysis

  • XFOILMIT · Drela

    Subsonic airfoil analysis and polars

  • OpenVSPNASA

    Parametric aircraft geometry

  • OpenAI

    Mission parsing and tool orchestration

  • Plotly

    Interactive engineering charts

  • FastAPI

    Python engineering & API backend

  • Next.js

    Web application frontend

Philosophy

Engineering First.
AI Second.

ERWIQ is an aerospace engineering platform that combines established engineering methods with AI orchestration. You describe a design requirement in plain language; ERWIQ plans the work and runs it through a sequence of engineering tools — conceptual sizing, airfoil selection, aerodynamic and stability estimates, propulsion, geometry — then presents the results with charts, a full narrative, and a structured PDF report.

It's built for aerospace engineers, UAV teams, and researchers who want automation without giving up scrutiny. The core principle: every number is traceable. Results come from named tool calls — never from a language model inventing values — and when a design is refined, each section is labeled as recomputed or carried forward, so you can always see what changed and what didn't.

Vision

The future of aircraft design is collaborative.

The interesting question isn't whether AI can design an aircraft on its own — it's how much further an engineer can go when the busywork disappears. ERWIQ exists to shorten the distance between a requirement and a concept you can scrutinize: the orchestration is automated, the record stays open, and the judgment calls stay where they belong — with the engineer.

That division of labor is the direction, and it doesn't change as ERWIQ grows: engineers set the intent, challenge the results, and make the decisions. ERWIQ runs the sequence, keeps the trace, and shows its work. Acceleration, not replacement.

Start with a mission.

Describe the requirement in plain language — ERWIQ runs the sequence and shows its work.

Launch App