Attic Engineering Internship

About Attic Research

We're a small, fast-moving team of engineers, hackers, and applied researchers. We figure out how complex systems work, where they break, and what to do about it. Interns here don't just shadow, they build, test, and contribute. We are looking for interns to join our team for full time roles in Summer of 2027 or year-round part time.

Available Internship Focus Areas

We have work that spans across multiple engineering disciplines, often overlapping in many cases. We are currently looking for creative and motivated students with interest in one or more of the following areas.

Cyber Physical Engineering

Reverse engineering, vulnerability research, and emulation

Work at the lowest levels of software and hardware. Figure out how a system works, then find where it breaks.

• Software and firmware reverse engineering in Ghidra, IDA Pro, and similar tools
• Vulnerability research on complex, non-traditional systems
• Firmware rehosting and full-system simulation with QEMU and custom emulators
• Hardware reverse engineering, embedded systems debugging, and in-situ instrumentation
• Custom hardware and software tools that automate the analysis workflow

Helpful background: Assembly, debuggers like GDB or x64dbg, embedded or firmware projects, debug interfaces (JTAG, SWD, UART), participation in CTF events.

Cognitive Radar Engineering

AI-powered detection and tracking on real hardware

Build algorithms that help radars pick targets out of clutter and keep track of them. Then prove them on real radar hardware, not just in simulation.

• Machine learning models that detect and classify targets in messy, real-world radar data
• Tracking algorithms that pair classical filters with learned models
• Signal processing across the full chain, from raw I/Q to tracks
• Hardware-in-the-loop testing that moves algorithms out of simulation

Helpful background: DSP fundamentals, radar or RF coursework, Kalman filtering or estimation theory, PyTorch, software-defined radios.

Counter-AI Engineering

Model reverse engineering, red-teaming, and real-world disruption

Treat AI models like any other system we analyze. Take them apart, find where they fail, and prove it outside the lab.

• Developing adversarial techniques to reverse engineer closed-weight models from the outside
• Red-teaming existing models to find and characterize their weaknesses
• Deploying real-world prototypes that disrupt model behavior, including in autonomous systems

Helpful background: deep learning fundamentals, adversarial ML or LLM red-teaming, deploying models on edge hardware.

Required for All Tracks

• Pursuing a B.S. or M.S. in Computer Engineering, Computer Science, Electrical Engineering, Mathematics, Physics, or a related field
• Familiarity with Python, C, or C++
• Strong interest in at least one of the tracks above
• Willingness to dig into open-ended problems until you understand them
• U.S. citizenship and the ability to obtain and maintain a U.S. Government security clearance

What You'll Gain

• Hands-on experience with high-impact, real-world systems
• Exposure to startup culture with direct access to subject matter experts
• Opportunity to work on projects that doesn’t just go into a slide deck, but actually make a difference
• Mentorship, flexibility, and the potential to return for full-time opportunities