Machine Learning Researcher Intern
Are you applying to the internship?
Job Description
About the Company:
Threat Detection and Response is a group within Cisco comprised of former security startup employees. They maintain a fast-paced, agile, and fun work environment. The team operates internationally, with the Prague, Czech Republic office collaborating closely with teams in the US. They strive for a balance of collaboration and autonomy, aiming to deliver highly effective solutions. Cisco’s established presence in the industry, combined with the team’s innovative approach to security, is presented as a unique advantage.
Job Description: Machine Learning Researcher Intern
This is a 20-hour-per-week hybrid internship based in the Prague (PRG5) office with an application deadline of April 30th, 2025.
Responsibilities:
• Assist in the entire machine learning model lifecycle: invention, design, development, and evaluation using large-scale data.
• Support the creation and implementation of efficient and scalable automated machine learning pipelines.
• Contribute to innovative projects from conception to execution under mentorship.
• Collaborate effectively with software engineers and clients to understand project needs.
• Demonstrate a strong eagerness to learn about cybersecurity data analytics and relevant tools.
• Effectively communicate findings through internal reports, blogs, and assist in preparing scientific papers and conference presentations.
Possible Projects:**
• Similarity Search on Sequence of Endpoint Security Events: Analyze cybersecurity event streams from network devices to identify similarly behaving devices using similarity search techniques. This involves segmenting data streams, creating embeddings, and finding similar sequences.
• Advance Command Line Preprocessing: Develop a system to correlate cybersecurity events into incidents by analyzing command line activity on affected devices. The focus is on normalizing command lines, removing or masking sensitive parameters (usernames, hashes, etc.) to reveal underlying patterns.
• Learning on Cybersecurity Relational Data: Explore methods to leverage relationships between various cybersecurity entities (IP addresses, URLs, emails, etc.) for automated threat identification and analyst support. This involves working with large graph data structures and potentially incorporating tabular, textual, and unstructured data.
Minimum Requirements:
• Currently pursuing a degree in Computer Science, Data Science, Machine Learning, or a related field.
• Proficiency in English.
Preferred Requirements:
• Basic understanding of software engineering principles.
• Coursework in one or more machine learning areas: supervised, semi-supervised, or unsupervised learning; model explainability; advanced statistics; graph theory; game theory.
• Strong problem-solving and logical reasoning skills.
• Enthusiasm for project completion and success.
• Hands-on experience with scikit-learn, Pandas, PyTorch, or (py)Spark.
