Jahan Khan

Jahan Khan · Technical portfolio for Anduril

Software &
Embedded SystemsJahan Khan · Early-career roles & internships

A portfolio of aircraft-design software, embedded robotics, flight-software security research, and operational systems.

Why AndurilAnduril is where I want to build my career. Learning from mentors at Anduril has strengthened that ambition and shaped the kind of engineer I want to become. I believe in the mission of strengthening American and allied defense, and I want to contribute to it through work I can take responsibility for. I’m looking for the opportunity to keep learning from people I respect while earning their trust through what I build.

Embedded systems[FIRMWARE · FLIGHT SOFTWARE]
Bare-metal ARM firmware, custom control boards, CAN, and a ROS 2 autonomy stack carried through field trials. Built to be tested, then handed to the people who run it.
IbriumRoverAircraft designFlight-software analysis
Download résuméExplore the engineering

U.S. citizen · ITAR-eligible · Irvine, California
Available now for full-time onsite work in Costa Mesa, or an internship alongside coursework.

Selected work

Engineering from requirement to delivery.

What I built, how it works, and where I took responsibility.

O3 aircraft program · February 2026–present

AI-Assisted Aircraft Design

Engineering software that connects aircraft design, analysis, and validation in one repeatable workflow.

I built the design tool alongside O3’s aerospace engineers to help them investigate design changes without losing the evidence behind each result. My work connected specialist analysis tools, automated repeated evaluations, and made the results easier to compare and review.

Program context: blended-wing-body design software for an attritable unmanned aircraft.

I worked with aerospace engineers to turn the coupled aerodynamic, structural, and aeroelastic behavior of a blended-wing body into a repeatable Python workflow across OpenVSP and CalculiX. Validation records preserve artifacts at the handoffs between tools and flag uncertain results for engineering review.

Relevant experience
Aeroelastic analysis learned from the engineers who do it and turned into working software
Engineering contribution
Validating results against published references before anyone relies on them

TechnologyPython · NumPy · SciPy · OpenVSP · CalculiX

Analysis chain
  1. IntegrationPython, NumPy, SciPy, OpenVSP, and CalculiX from parametric geometry through aerodynamic, structural, mass, and aeroelastic analysis.
  2. EvaluationHundreds of variants screened; three finalists evaluated with higher-fidelity tools.
  3. ResultA trade-study design with approximately 24% lower structural mass than its baseline.

Ibrium Studios · September 2025–May 2026

Distributed Motor Control on CAN

Production delivery
Embedded Systems Engineer · Disney contract

Synchronized multi-axis motion across a central controller and embedded actuator nodes, delivered on a fixed production date.

I designed the multilayer motor-control PCBs in Altium, wrote the bare-metal ARM Cortex-M node firmware, and defined the CAN protocol that kept the nodes in step with the central controller. My responsibility ran from schematic and board bring-up through integrated motion, fault handling, operating procedures, and training the crew who ran it without me. I also integrated SLAM-based registration that aligned the physical set to its Unreal Engine model in real time, so the motion system tracked the environment it stood in.

Relevant experience
Bare-metal ARM Cortex-M firmware, CAN, and custom motor-control boards
Engineering contribution
Board bring-up and debugging with logic analyzers and oscilloscopes on live hardware

TechnologyC · ARM Cortex-M · CAN · UART · SPI · I2C · PWM · Altium · Oscilloscopes · Logic analyzers

Boards
Multilayer motor-control PCBs, schematic through fabrication and board bring-up.
Firmware
Bare-metal control loops, feedback, fault detection, and safe-stop on each node.
Bus
A CAN protocol and central controller for synchronized real-time motion across nodes.
Delivery
Revalidated motion, documented procedures, and trained production operators under a fixed deadline.
Engineering record · the CAN-control latency fault
  1. ObservedIntermittent control latency during integrated multi-axis motion.
  2. InvestigatedLogic-analyzer timing traces on the bus isolated a control-loop deadline overrun in the node firmware.
  3. FixedMoved control scheduling to a fixed-rate timer interrupt.
  4. ValidatedRe-ran the integrated motion sequence, then wrote the operating procedures and trained the crew.

Saddleback College Robotics · 2021–2023

Rover Autonomy and Ground Control

Two competition seasons
Co-Software Lead · University Rover Challenge

A ROS 2 autonomy stack connecting navigation, sensor fusion, motor control, and operator tools through rover field trials.

I co-led the ROS 2 autonomy stack: GNSS waypoint following, IMU and stereo-camera fusion, LiDAR obstacle avoidance, and Nav2 behavior trees. An asynchronous Rust driver spoke COBS-framed postcard messages to STM32 boards; ODrive motor controllers were commanded over CAN with the drive model tuned in MATLAB/Simulink. A React/TypeScript and Foxglove ground station carried telemetry, waypoint commands, and a Jetson H.264 video link over the radio.

Relevant experience
Sensor integration, STM32 control interfaces, simulation, and recorded-run replay
Engineering contribution
Field testing autonomy, control, and telemetry on hardware, then fixing what failed

TechnologyC++ · Rust · Python · ROS 2 / DDS · Nav2 · STM32 · CAN · GNSS · IMU · Gazebo · Isaac Sim · rosbag · Docker · MATLAB/Simulink

Integrated hardware
GNSS, IMU, cameras, LiDAR, ODrive motor controllers, and serial links to STM32 subsystems.
Test without the vehicle
Gazebo and ros2_control, Isaac Sim, rosbag replay of field runs, and a Docker-based test harness.
Telemetry
Jetson H.264 video, telemetry, and waypoint commands connected to the ground-control interface.
Team
Coordinated subsystem integration through GitHub reviews across drive, arm, science, and autonomy.
From sensor inputs to testable behavior
  1. ObserveCamera, LiDAR, GNSS, and IMU inputs support perception and state estimation.
  2. PlanNav2 behavior trees, waypoint navigation, and obstacle avoidance produce movement decisions.
  3. ControlRust messaging and STM32 interfaces connect commands to motor-control hardware.
  4. EvaluateSimulation and rosbag replay help investigate behavior and test software changes.

Software and validation

Software I built to be tested.

Source analysis and signal processing: tools that make software behavior easier to investigate and evaluate.

Self-directed curriculum project · 2026

Flight-Software Static Analysis (PX4)

Flight-software analysis · Rust / PX4 v1.16

I built exploit-discovery tooling to study flight software and identify code patterns for security review.

The Rust analyzer turns C/C++ source into a searchable view of functions and their direct calls. I used PX4 v1.16 to study how MAVLink command handlers decode messages and pass work to other functions, then added heuristic rules that identify code worth closer review.

Source → structure → review
  1. ExtractDiscover function definitions and direct call sites, distinguishing resolved names from unknown targets.
  2. InspectLook up a handler such as handle_message_command_int and inspect its direct callees.
  3. PrioritizeFlag missing authentication references, unchecked memory-access patterns, and unexpected navigation-state writes.
Engineering depth
Rust command-line tooling, C/C++ source structure, call-graph extraction, and MAVLink command handling in PX4.
Validation
Recorded validation: 29 tests passed, the release build succeeded, and handler lookups were checked against PX4 source. Vulnerable and corrected fixtures exercised rule behavior.
Review model
Findings identify source patterns for investigation. Function lookup reports direct calls; unresolved targets remain visible rather than being treated as proven call paths.

TechnologyRust · C/C++ source analysis · PX4 v1.16 · MAVLink · Call graphs · Unit tests · CLI tooling

Independent · March 2026–present

Counter-UAS Electronic-Warfare Decision Engine

Haskell · simulation and replay

A Haskell signal-processing and decision pipeline turns synthetic or SigMF replay IQ into emitter observations and constrained response comparisons, and keeps the evidence behind each plan so behavior can be investigated against the inputs that produced it.

Execution path
  1. Signal processingRadix-2 FFT, CA-CFAR detection with configurable guard and training cells, emitter-feature extraction, threat-library matching.
  2. Decision logicExplicit mission state, spectrum and resource checks, and confidence-based abstention when detection evidence is weak.
  3. ValidationAdversarial scenarios, plan-scoped records, replay evidence, and reproducible Nix environments.
Relevant experience
RF signal processing from IQ to decision, built and tested alone
Engineering contribution
Recorded inputs and replay so every change can be checked against what the system saw

TechnologyHaskell · Go · TypeScript · Python · Nix

Field and operational experience

Software in its working environment.

O3 · March 2026–present

Shipyard Operations Software

Deployed software · HII shipbuilding use case

Built and deployed shipyard-operations software for O3 in Rust and Python: a PostgreSQL ontology of parts, equipment, workers, and task dependencies, OPC-UA telemetry adapters, and OR-Tools CP-SAT scheduling against resource and task constraints.

Packaged for local deployment with Docker, with role-based access, event history, and tamper-evident audit records so operational changes stay reviewable.

Relevant experience
Software engineering in Rust and Python beyond firmware
Engineering contribution
Adapting software to operator requirements from direct feedback

TechnologyRust · Python · FastAPI · PostgreSQL · React · TypeScript · OPC-UA · OR-Tools · Docker

Tata · December 2025–February 2026

Field Researcher and Developer

Apple AR/LiDAR research partnership

Conducted AR/LiDAR research across field sites, traveling throughout the engagement to configure sensors and validate captures. Developed C software and repeatable data-collection protocols for consistent indoor scanning.

Relevant experience
Sensor configuration and capture validation in the field
Engineering contribution
Repeatable data-collection procedures other people can run

TechnologyC · AR/LiDAR · Sensor configuration · Field validation

MedBWS · May 2020–June 2023

Automation Lead

Healthcare software deployment

Built a computer-vision and Swift iOS workflow for pharmaceutical identification and disposal, deployed across hospitals in a HIPAA-compliant environment, with authentication, authorization, secure APIs, and audit logging for controlled-medication records.

Relevant experience
Software deployment for operational users
Engineering contribution
Documented procedures and audit records in a regulated setting

TechnologyComputer vision · Swift / iOS · Secure APIs · Authorization · Audit logging

Capabilities

What I have built so far.

Grouped by the three areas I want to keep learning in at Anduril.

Firmware & hardware

[FIRMWARE FOR FIELDED HARDWARE]

Cores
ARM Cortex-M, STM32
Firmware
Bare-metal, RTOS, C
Buses
CAN, UART, SPI, I2C, PWM
Hardware
Altium PCB design, board bring-up
Debug
JTAG/SWD, oscilloscope, logic analyzer

Flight software & autonomy

[AUTONOMY FOR VEHICLES]

Languages
C++, Rust, Python
Stack
ROS 2, Nav2, behavior trees
Sensors
GNSS, IMU, cameras, LiDAR fusion
Control
ODrive motor control over CAN
Flight software
PX4 / MAVLink source analysis

Test & simulation

[TESTING BEFORE HARDWARE]

Simulation
Gazebo, Isaac Sim, ros2_control
Replay
rosbag field-run replay
Bench test
Docker harness that runs the full stack with no vehicle attached
Builds
Nix reproducible environments
Verification
Unit and fixture tests, validated builds
Evidence
PX4 analysis

Engineering capabilities

Where I can contribute.

A direct path from software and embedded-systems responsibilities to the work behind them.

Firmware

Bare-metal firmware and real-time control

Nine months of bare-metal ARM Cortex-M node firmware under a production contract, plus two seasons of STM32 firmware on the rover.

Ibrium Studios Firmware

Bus protocols: SPI, I2C, UART, CAN

A CAN protocol I defined for synchronized motor nodes; UART, SPI, I2C, and PWM in the node firmware; CAN drive control on the rover.

Ibrium Studios Hardware

Hardware/software debugging

The CAN-control latency fault: logic-analyzer traces to a control-loop deadline overrun, fixed in firmware and revalidated.

Engineering record Flight software

Sensor integration and autonomy validation

GNSS and IMU fusion, STM32 interfaces through a Rust serial driver, Gazebo simulation, and replay of recorded field runs.

Saddleback Robotics Flight software

Field robotics and operator interfaces

Two University Rover Challenge seasons as co-software lead: ODrive motor control over CAN, a ground station, and an H.264 telemetry link over the base-station radio.

Saddleback Robotics Flight software

PX4 and MAVLink familiarity

Flight-software analysis and exploit discovery: Rust tooling for C/C++ function discovery, direct-call inspection, and heuristic review of MAVLink handlers in PX4 v1.16.

Flight-software projectOwnership

Signal processing and constrained decisions

An RF signal-processing and decision engine built from SigMF replay data, validated in adversarial scenarios, with reproducible Nix builds.

EW decision engine Flight software

Low-level drivers on microcontrollers and application processors; C++ and Rust

An asynchronous Rust driver speaking COBS-framed postcard messages to STM32 boards, alongside C++ autonomy nodes on ROS 2.

Saddleback Robotics Test infrastructure

Simulation, replay, and reproducible environments

Gazebo and Isaac Sim scenes, rosbag field-run replay, Docker test tooling, and Nix-pinned environments for the EW engine.

Saddleback Robotics

I am looking for early-career software, embedded systems, and integration roles where I can learn from experienced engineers. Each row links to the work behind it.

Contact

Jahan Khan

Available now for full-time onsite work in Costa Mesa, or an internship alongside coursework.

jahanzebkhan214@gmail.com(949) 872-0649Download résumé

Academic foundation · Irvine Valley College

Relevant coursework

Computer Engineering · All A grades

  • C++ programming
  • Computer Organization and Assembly
  • MATLAB
  • Python
  • Calculus
  • Linear Algebra
  • Differential Equations

CodeSignal Industry Coding Assessment

600/600

August 2026

Citizenship

U.S. citizen · ITAR-eligible

Eligible for export-controlled programs.

Letters of recommendation and code samples available upon request.