Gopi Naga Vignesh Thirupati

Aerospace & Computational Simulation Software Engineer

3-DoF/6-DoF Dynamics • Multi-Stage Rocket Staging • SGP4 Orbit Propagation • Physics-Informed CFD • AI Model Evaluation (SWE-bench) • Docker & MCP

I am an Aerospace & Computational Simulation Software Engineer specializing in translating complex physical dynamics, multi-stage rocket trajectories, orbital mechanics, and fluid flows into high-performance, test-driven software, containerized microservices, and Model Context Protocol (MCP) tools. My work enables autonomous AI systems (Claude, Cursor, Antigravity) to reason through and simulate atmospheric flight physics, satellite orbits, aerodynamic surfaces, and propulsion thermodynamics with deterministic precision.

• 🚀 Core Focus: Computational flight simulation, multi-stage vehicle staging, physics-informed AI surrogates, cloud-native aerospace microservices, and reproducible AI coding benchmarks (SWE-bench).
• 🛠️ Engineering Stack: Python (NumPy, SciPy, Matplotlib), C++, TypeScript, Pytest (140+ passing tests), Docker, Docker Compose, GitHub Actions (CI/CD), Linux.
• 📐 Domain Expertise: 1976 U.S. Standard Atmosphere (0–86 km), Transonic Drag Divergence, Runge-Kutta (RK4) Solvers, SGP4/J2 Secular Perturbations, Hohmann Orbit Transfers, Vortex Panel Methods, Squire-Young Viscous Drag, NACA 4/5-Digit Camber, Compressible Isentropic Flow & Shock Waves, Tsiolkovsky Staged Rocket Propulsion.
• 🤖 AI Benchmarking & Tooling: Designing deterministic fail-to-pass test harnesses and golden patches for LLM code evaluation across physical ODEs and aerodynamic singularities.

graph TD
    subgraph Core Physics & Astrodynamics Engines
        AFD["🚀 aero-flight-dynamics<br/>(3-DoF RK4 & Multi-Stage Staging)"]
        CFD["🌪️ cfd-airfoil-surrogate<br/>(Vortex Panel, NACA 4/5-Digit & Cp)"]
        OTA["🛰️ orbital-telemetry-agent<br/>(SGP4/J2 Orbit & Hohmann Transfer)"]
        OET["📐 openengineering-toolkit<br/>(Propulsion, Compressible & Mechanics)"]
    end

    subgraph AI Agent & Evaluation Harness
        AEB["🔬 aero-eval-benchmark<br/>(SWE-bench Style AI Coding Benchmark)"]
        MCP["🤖 Model Context Protocol (MCP)<br/>(Claude Desktop • Cursor • Antigravity)"]
    end

    subgraph Cloud-Native Production Stack
        OPS["🌐 aerospace-ops-stack<br/>(FastAPI Microservices & Mission Control UI)"]
        DOCKER["🐳 Docker & Docker Compose<br/>(Production Container Network: aero-mesh)"]
    end

    AFD -->|JSON-RPC 2.0| MCP
    CFD -->|JSON-RPC 2.0| MCP
    OTA -->|JSON-RPC 2.0| MCP
    OET -->|JSON-RPC 2.0| MCP
    MCP --> AEB
    AFD --> OPS
    CFD --> OPS
    OTA --> OPS
    OET --> OPS
    OPS --> DOCKER

Cloud-Native Aerospace Microservices Stack & Mission Control Telemetry Visualizer with Docker Compose
Python • Docker Compose • FastAPI • HTML5 Canvas • CI/CD Matrix • 10 Automated Tests • MIT Licensed

• Central Mission Gateway: Microservice hub orchestrating 3-DoF suborbital flight simulations, real-time SGP4 orbital propagation, Hohmann delta-v budgets, compressible gas dynamics, and SWE-bench AI benchmark evaluation.
• Interactive Mission Control UI: Real-time canvas telemetry gauges (altitude, velocity, Mach, Max-Q), dynamic orbital ground tracks with elevation footprint circles, and airfoil surface pressure distribution charts.
• Docker Compose Architecture: Production container network (aero-mesh) with healthchecks, isolated service bridges, and multi-stage builds.

Autonomous Evaluation Harness & SWE-bench Style Benchmark for Testing Coding LLMs on Aerospace & Physics Tasks
Python • SWE-bench Protocol • Docker • MCP Server • Multi-Python CI Matrix • 10 Automated Tests • MIT Licensed

• SWE-bench Standard Architecture: Curated benchmark challenge matrix testing frontier LLMs on numerical singularities, transonic wave drag discontinuities, and orbital coordinate branch cuts.
• Isolated Evaluation Harness: Validates unified diff patches against rigorous fail-to-pass edge case tests while enforcing zero regressions on pass-to-pass baselines.
• Frontier Model Leaderboard: Benchmarked performance data for Claude 3.5 Sonnet, GPT-4o, DeepSeek-Coder, and Qwen-2.5-Coder on physical ODEs.
• MCP Integration: Exposes aeroeval_list_tasks, aeroeval_get_task_spec, and aeroeval_evaluate_patch for autonomous AI agents.

Physics-Informed CFD Aerodynamic Surrogate Model for Instantaneous Pressure ($C_p$), 5-Digit Geometry & Polars
Python • Docker • Docker Compose • MCP Server • UIUC Wind Tunnel Validation • 18 Automated Tests • MIT Licensed

• Instantaneous Prediction (<2 ms): Generates chordwise surface pressure distributions ($C_p(x/c)$) and aerodynamic polars ($C_L, C_D, C_M, L/D$) 23,000x faster than 2D OpenFOAM RANS.
• NACA 4 & 5-Digit Airfoils: Parametric cubic camber generator supporting high-performance 5-digit airfoils (e.g. NACA 23012) and 4-digit profiles.
• Pitching Moment & Center of Pressure: Quarter-chord moment ($C_{m, c/4}$), leading-edge moment ($C_{m, LE}$), and center of pressure ($x_{cp} / c$) stability modeling.
• Vortex Panel & Boundary Layer: Cosine leading-edge clustering, Kutta closure, and Squire-Young profile drag with stall roll-off.
• Tooling: Native aeroml CLI, Docker container, and MCP server for direct AI tool calling.

Autonomous Satellite Orbit Propagator, Hohmann Transfer Calculator & Model Context Protocol (MCP) Server
Python • Docker • Docker Compose • Multi-Python CI Matrix • 15 Automated Tests • MIT Licensed

• Astrodynamics & J2 Perturbation: Keplerian orbit propagator modeling Earth oblateness ($J_2 = 1.08263 \times 10^{-3}$), nodal regression ($\dot{\Omega}$), and argument of perigee precession ($\dot{\omega}$).
• Orbital Transfers & Maneuvers: Two-impulse Hohmann transfer delta-v budgets ($\Delta v_1, \Delta v_2$), transfer duration ($t_{\text{tof}}$), and combined apogee plane change optimization.
• Ground Station Tracking: Topocentric SEZ coordinate transformation predicting satellite Acquisition of Signal (AOS), Loss of Signal (LOS), max elevation, and pass schedules.
• RF Doppler Telemetry: Real-time line-of-sight range rate and RF Doppler frequency shift calculations for satellite radios.
• MCP Server Integration: Native JSON-RPC 2.0 tools for Claude Desktop and Cursor (propagate_orbit, predict_ground_passes, calculate_doppler_shift, calculate_hohmann_transfer).

High-Fidelity 3-DoF Aerospace Simulation Engine, Multi-Stage Staging & Model Context Protocol (MCP) Server
Python • Docker • Docker Compose • Multi-Python CI Matrix • 35 Automated Tests • MIT Licensed

• Multi-Stage Vehicle Dynamics: Discrete stage separation, dry mass casing jettison, inter-stage coast, and sustainer ignition modeling.
• MCP Server Integration: Exposes JSON-RPC 2.0 tools for Claude Desktop, Cursor, and AI agents to simulate 3-DoF flights, staged trajectories, evaluate atmospheric layers, and analyze airfoils.
• Atmospheric Physics: Exact 7-layer 1976 U.S. Standard Atmosphere model up to 86,000 meters geopotential altitude with Sutherland dynamic viscosity law.
• Compressible Aerodynamics: Prandtl-Glauert subsonic compressibility corrections, Lock/Korn transonic wave drag divergence with non-linear saturation clamping ($C_{D,\text{wave}} \le 0.40$).
• Numerical Solver: 4th-order Runge-Kutta (RK4) integrator tracking altitude, Mach number, axial G-load, dynamic pressure (Max-Q), and telemetry data.
• Dual Recovery Subsystem: Staged drogue chute deployment at apogee and main canopy inflation for certified safe landing ($v_{\text{touchdown}} &lt; 6.0\text{ m/s}$).

Transparent, Verified Computational Mechanics, Compressible Gas Dynamics & Rocket Propulsion Library
Python • FastAPI • Docker • Docker Compose • CLI • MCP Server • 52 Automated Tests • MIT Licensed

• Compressible Gas Dynamics: Isentropic temperature and density ratios ($T_0/T, \rho_0/\rho$), nozzle sonic throat area ratio ($A/A^*$), Rankine-Hugoniot 1D normal shock jumps ($M_2, P_2/P_1, P_{02}/P_{01}$), and Prandtl-Meyer expansion angle ($\nu(M)$).
• Rocket Propulsion Analytics: Tsiolkovsky delta-v, characteristic exhaust velocity $c^*$, nozzle thrust coefficient $C_F$, and total thrust.
• Fluid Dynamics & Sizing: Reynolds number regime verification, dynamic pressure $q$, compressible Mach number, and Haaland pipe friction.
• Structural Solid Mechanics: Euler-Bernoulli mid-span beam deflection, bending stress, and factor of safety (FoS).
• Cross-Platform Access: Exposed via openeng CLI, FastAPI REST endpoints, containerized Uvicorn daemon, and native MCP stdio server.

• Email: vigneshthirupati2000@gmail.com
• LinkedIn: linkedin.com/in/vignesh375
• GitHub Profile: github.com/perfectalone375