Principal Engineer, AI Security & Model Internals

Location: Wrocław, Warsaw (Hybrid)

Most people treat language models as black boxes. At Opera, we want to understand what happens inside them.

We’re looking for a hands-on Principal Engineer to join our AI & Browser Innovation Hub in Wrocław, focusing on model internals and AI security. Working closely with Opera’s AI leadership, as a technical partner in evaluating and challenging engineering approaches, you’ll investigate how language models behave, identify vulnerabilities, and develop security solutions for AI-powered browsing systems used by millions of people.

This role combines applied research, adversarial testing, and engineering, with the opportunity to turn emerging techniques into practical solutions. You’ll have the autonomy to challenge assumptions, explore new approaches, and contribute directly to the security of Opera’s AI-powered products.

Role & Responsibilities

• Model internals. Build tooling to inspect open-weight models layer by layer and head by head: activation capture, linear probes, logit and tuned lens, activation patching, ablation and sparse autoencoders.

• Model supply chain. Scan third-party weights and fine-tunes for backdoors, trojans and hidden behaviour before they reach users, think heuristic antivirus, but for neural networks.

• Runtime monitoring for agents. Design monitors that read a model's internal state at critical moments, such as tool calls, and escalate to a second model for verification when something looks wrong.

• Prompt injection defence. Red-team our agentic browsing and harden it through model-level work, but above all through architecture that keeps untrusted web content away from privileged actions.

• Secure fine-tuning. Build fine-tuning pipelines (LoRA and beyond) that resist data poisoning, and test the results for memorisation, leakage and safety regressions.

• Research to product. Read the papers, reproduce the promising ones in days, and turn the ones that hold up into monitors and checks that run within the constraints of on-device inference.

• Share what you find. Write it up, open-source the tools where it makes sense, and help build Opera's voice in AI security.

Job Requirements

• Strong engineering in Python and PyTorch, with a track record of building real systems, not only notebooks.

• Hands-on experience with transformer internals. You can explain the residual stream, and you've hooked into a model's forward pass yourself, whether with TransformerLens, nnsight or raw hooks.

• Practical fine-tuning experience (LoRA/QLoRA, PEFT) and the habit of evaluating what you trained.

• An adversarial mindset you can show evidence of: red-teaming, vulnerability research, reverse engineering, CTFs, jailbreak work or something comparable. We care about how you think about breaking things more than about security titles.

• The ability to take an idea from paper to working prototype quickly, and to tell the difference between a real finding and an artefact.

• Clear written English, and the confidence to push back on ideas, including your manager's.

Nice to Have

• Mechanistic interpretability work: sparse autoencoders, circuit analysis, probing, published or open-source.

• ML security research: backdoor detection, data poisoning, model extraction, membership inference.

• Inference runtimes and performance: llama.cpp, vLLM, MLX, CUDA or Metal kernels, quantisation.

• Browser security, sandboxing or agent frameworks.

• C++/Rust for integrating monitors into native code paths.

• A history of taking things apart for fun game trainers, memory editors, emulators, firmware.

What’s On Offer

At Opera, you’ll join a diverse and inclusive team of experienced, supportive professionals who value creativity and collaboration. We work in a flat structure with short decision-making paths, use smart technology, and support your ongoing skill development in a friendly and empowering environment.

Interested?

We’d love to hear from you! Applications are reviewed on a rolling basis, so we encourage you to apply soon. Please submit your CV in English. Have questions about our recruitment process, remote work, or benefits? Check out our FAQ page for more details.