Minor Computer Science

1. Cryptography & Web Technologies

A browser extension that demonstrates cryptographic hash functions (e.g., SHA-256, MD5, SHA-3) in real time as a user types. It helps users understand the avalanche effect of hashes by showing how a single character change alters the output drastically. This tool can also explain properties like collision resistance, pre-image resistance, and digest length. Its educational focus makes cryptography concepts more accessible to developers, students, and security enthusiasts. Such visualization bridges the gap between abstract cryptographic theory and practical application in modern web technologies.

2. Network Security & Operating Systems

A FUSE-based user-space file system that enforces encryption selectively at the process or user level. Unlike traditional full-disk encryption, this approach applies the principle of least privilege by encrypting files only when accessed by specific applications. It minimizes exposure of sensitive data while ensuring seamless user experience. This design leverages OS-level security primitives to isolate file operations, preventing unauthorized processes from reading confidential files. Such a system is particularly useful in multi-user or containerized environments.

3. Machine Learning & Network Management

An anomaly detection framework that applies unsupervised learning to live network traffic data. By clustering flows (e.g., via k-means or DBSCAN), the system identifies deviations from normal traffic patterns without requiring labeled datasets. Detected anomalies can represent potential intrusions, misconfigurations, or DDoS activity. Alerts are raised to administrators with metadata such as flow source, destination, and traffic volume. This improves proactive monitoring in dynamic, large-scale networks where supervised signatures fail to adapt.

4. Data Engineering & Web Technologies

A drag-and-drop ETL (Extract, Transform, Load) pipeline builder with a web-based interface. Users can design workflows that connect multiple data sources, apply transformation logic, and define data sinks visually. The system automatically generates optimized ETL execution scripts from the design. It allows real-time pipeline validation and performance estimation. Integration with modern data platforms (e.g., Spark, Flink, or cloud-native services) ensures scalability. Such a tool simplifies data engineering tasks for non-programmers while maintaining technical depth for experts.

5. Operating Systems & Software Engineering

A custom command-line shell that embeds lightweight version control mechanisms. Inspired by Git, it enables developers to snapshot, diff, and rollback configuration files or scripts directly from the shell. The system manages file states transparently, ensuring resilience against accidental misconfigurations. It incorporates OS-level primitives for efficient file tracking and metadata storage. This tool provides developers with a hybrid environment for both execution and configuration management. It is particularly useful in DevOps and systems administration workflows.

6. Cryptography & Machine Learning

A privacy-preserving ML inference system based on homomorphic encryption. The model (e.g., logistic regression) operates directly on encrypted user data without requiring decryption. Users can outsource computation to cloud servers without exposing sensitive inputs. While computationally intensive, this approach maintains confidentiality in healthcare, finance, or biometrics applications. It showcases the integration of secure cryptographic primitives with AI-driven analytics. The system demonstrates how encrypted computations can balance usability and security in modern ML deployments.

7. Network Security & Data Engineering

A distributed log aggregation system designed to securely collect and process logs from multiple devices across a network. It uses TLS or mutual authentication to ensure trusted log ingestion. The logs are stored in a scalable data pipeline where filters and rules can be applied in real time. Administrators can configure alerts on suspicious activity or policy violations. This system enables forensic analysis by correlating distributed logs in a unified manner. It is highly useful in SOC (Security Operations Center) environments.

8. Web Technologies & Machine Learning

A recommendation engine that applies graph-based algorithms to suggest content. By modeling users and items as nodes in a graph, the system detects communities or related clusters. This allows personalized recommendations that capture deeper contextual relationships than simple collaborative filtering. The tool integrates with blogs or e-commerce platforms via APIs. It updates in real time as new interactions occur, improving relevance. This graph-based approach enhances user engagement through intelligent navigation of related content.

9. Software Engineering & Operating Systems

A distributed build system for C++ projects that automatically resolves dependencies. It leverages a distributed hash table (DHT) to cache build artifacts across multiple machines. This reduces redundant compilation and accelerates build times in large-scale projects. The system integrates with existing toolchains (e.g., CMake, Make) while providing platform independence. It also ensures reproducibility by tracking versioned dependencies. Such automation streamlines complex software development lifecycles.

10. Network Management & Cryptography

A secure key distribution and management system that uses a lightweight, custom protocol. It provides cryptographic keys across distributed nodes while preventing interception or replay. The protocol ensures forward secrecy and periodic key rotation. It integrates with existing network management frameworks to minimize overhead. Such a system is crucial for IoT or constrained environments where traditional PKI is too heavy. This ensures secure communication without centralized bottlenecks.

11. Data Engineering & Network Security

A real-time intrusion detection system (IDS) built on a stream processing framework such as Apache Flink. Incoming packets are ingested and analyzed on the fly, enabling detection of malicious payloads within milliseconds. Rule-based and pattern-matching mechanisms identify threats while machine learning models can adapt to new attacks. The streaming architecture ensures scalability in high-bandwidth networks. It reduces the delay between attack detection and mitigation. This approach strengthens enterprise network resilience.

12. Machine Learning & Operating Systems

A reinforcement learning–based process scheduler that dynamically adjusts priorities. Instead of static heuristics, it learns from past system behavior to optimize resource allocation. The scheduler balances throughput, fairness, and responsiveness based on workload patterns. It can adapt to user-centric metrics such as minimizing application latency. This enhances overall system performance, especially in multi-core and heterogeneous processor environments. Such a design introduces AI-driven decision-making into low-level OS components.

13. Web Technologies & Network Management

A real-time web dashboard that visualizes live network topology. It uses SNMP to gather device information and WebSockets for push-based updates. The dashboard shows node status, bandwidth utilization, and fault alerts on an interactive map. Network administrators can drill down into device metrics and configure alerts. Its browser-based nature makes it accessible across devices without extra software. This tool modernizes network management with responsive visualization.

14. Cryptography & Software Engineering

A static analysis tool tailored to cryptographic implementation. It scans codebases to detect common vulnerabilities such as weak random number generation, improper key storage, or unsafe algorithm modes (e.g., ECB in AES). The tool integrates with CI/CD pipelines for automated security checks. It can suggest best practices based on NIST or OWASP standards. By preventing subtle cryptographic errors at development time, it reduces downstream security risks. This makes it invaluable in secure software development workflows.

15. Operating Systems & Data Engineering

A file system that integrates data deduplication and hot-block mapping. It identifies redundant data at the block level and stores only unique content, saving disk space. Simultaneously, it maintains metadata to track access frequency and generates a “hot map” of frequently used blocks. This provides insights into performance bottlenecks or caching opportunities. The design helps optimize storage efficiency and I/O throughput. It is highly applicable in enterprise storage and cloud systems.

16. Network Security & Web Technologies

A decentralized multi-factor authentication (MFA) system for web applications. Instead of relying on a central server, it uses QR codes and a mobile app for verification. The QR code encodes a cryptographic challenge, which the mobile app signs and returns. This removes dependency on traditional username-password models. It also enhances resistance against phishing and centralized breaches. Such MFA is well-suited for high-security environments requiring lightweight, distributed solutions.

17. Machine Learning & Cryptography

A federated learning system where multiple parties collaboratively train a model without sharing raw data. Secure aggregation ensures only encrypted model updates are exchanged. This approach preserves privacy while still allowing global model convergence. It is suitable for sensitive domains like healthcare or finance where data cannot leave local premises. The system leverages cryptographic protocols such as secret sharing and secure multiparty computation. It balances accuracy with privacy and regulatory compliance.

18. Data Engineering & Software Engineering

A framework for generating synthetic test data from database schemas. Rules define realistic ranges, dependencies, and constraints. The synthetic data preserves statistical properties without exposing sensitive information. It is useful for testing pipelines, load testing, and compliance with data privacy regulations. The framework supports integration with CI/CD pipelines for automated test deployments. By ensuring data variety, it strengthens software robustness and debugging.

19. Network Management & Operating Systems

A custom DNS server that enforces security and content policies. It can block malicious domains and apply user-defined filtering rules. The server integrates with threat intelligence feeds for real-time updates. Administrators can define blacklists, whitelists, and parental control policies. It operates efficiently across distributed environments while maintaining caching for performance. Such a system enhances both security and control at the DNS layer.

20. Web Technologies & Operating Systems

A web-based IDE running on a sandboxed virtual machine. The VM isolates user code execution, ensuring system security. The IDE supports multiple languages and provides debugging, code completion, and file management. It enables collaborative coding through browser-based access. Built-in virtualization ensures users cannot escape the sandbox. This approach democratizes software development environments and enhances accessibility.

21. Cryptography & Data Engineering

A secure data warehouse using format-preserving encryption (FPE). Sensitive fields (like credit card numbers) remain searchable and analyzable without exposing plaintext. The system supports range queries and analytics while maintaining compliance with regulations (e.g., GDPR, HIPAA). It integrates with ETL pipelines to automatically encrypt data at ingestion. By combining data engineering efficiency with strong cryptography, it secures enterprise-scale datasets. This enables analytics without compromising confidentiality.

22. Network Security & Machine Learning

A deep learning system to classify network attacks based on packet data. Using CNNs or RNNs, it learns to distinguish DDoS, port scanning, or intrusion attempts. The model continuously adapts as new traffic patterns emerge. Real-time inference ensures threats are flagged before damage occurs. It complements rule-based IDS by covering previously unseen attacks. Such integration improves the accuracy of SOC defenses.

23. Web Technologies & Software Engineering

A collaborative framework for real-time whiteboards and editors. It uses CRDTs (Conflict-Free Replicated Data Types) or OT (Operational Transform) to handle concurrent edits. Offline synchronization ensures edits persist even during disconnection. Users can annotate, code, or draw simultaneously in a shared space. The system scales to large groups while preserving consistency. This empowers distributed teams with seamless collaboration.

24. Operating Systems & Network Management

A lightweight protocol stack tailored for IoT devices. It optimizes for low-power operation and intermittent connectivity. Unlike traditional TCP/IP stacks, it minimizes memory footprint and retry overhead. The stack supports essential security mechanisms despite its compactness. It ensures reliable message delivery in constrained environments. Such a system extends IoT device longevity and performance in sensor networks.

25. Data Engineering & Machine Learning

A churn prediction system analyzing user clickstream data. By treating activity logs as time-series data, it applies forecasting or sequence models (e.g., LSTMs). The system identifies early signals of disengagement, allowing interventions like personalized offers. It integrates into business dashboards to provide retention insights. This approach transforms raw user events into predictive intelligence. It helps companies reduce attrition and optimize user engagement.

26. Cryptography & Network Security

An encrypted messaging app with a custom elliptic curve–based key exchange protocol. Messages are encrypted end-to-end, ensuring only sender and receiver can read them. Keys are rotated periodically to reduce compromise risk. The custom exchange enhances flexibility while maintaining forward secrecy. The system is hardened against replay and man-in-the-middle attacks. It demonstrates secure communication across distributed endpoints.

27. Software Engineering & Network Management

An automated tool for network configuration using a declarative model. Like Ansible, configurations are described in code and applied consistently across devices. The system integrates with a central repository for version control. It ensures compliance by validating configurations before deployment. Automated rollback mechanisms restore previous states if failures occur. This reduces human error and accelerates infrastructure provisioning.

28. Machine Learning & Web Technologies

A web-based visualization tool that demonstrates the decision boundaries of classification algorithms such as logistic regression, SVM, random forests, and neural networks. Users can upload custom datasets to see how models behave with their data. The tool renders 2D/3D plots to show separation lines, hyperplanes, or clusters. It provides metrics like accuracy, precision, and recall in real time. A comparison mode allows users to test multiple algorithms side by side. This helps students and practitioners understand model behavior beyond abstract math. Its interactive interface bridges ML education and practical web deployment.

29. Operating Systems & Cryptography

A secure bootloader that verifies the integrity of the OS kernel and critical binaries. At boot time, it checks digital signatures against trusted root keys stored in hardware (e.g., TPM or UEFI Secure Boot). If verification fails, the boot halts, preventing malware persistence. It ensures the chain of trust from firmware to OS, protecting against rootkits. The system can also support remote attestation to prove device integrity to external servers. Such mechanisms are vital in enterprise and defense-grade environments. It combines low-level OS initialization with cryptographic enforcement.

30. Data Engineering & Network Security

A packet anonymization system that redacts sensitive fields such as IP addresses or MAC addresses while preserving traffic flow structure. It applies tokenization or pseudonymization so analysis remains valid. The anonymizer integrates with real-time stream frameworks (e.g., Apache Flink, Kafka Streams). It enables researchers to share traffic traces without violating privacy. The system supports reversible anonymization for trusted forensic users. By balancing privacy and utility, it fosters collaborative network security research.

31. Web Technologies & Cryptography

A client-side password manager that encrypts all secrets locally before syncing. Keys are derived from user passphrases with strong key-stretching (PBKDF2, Argon2). Data is synchronized peer-to-peer between devices using encrypted channels, avoiding central server dependency. Even if sync metadata leaks, attackers cannot recover passwords. The UI integrates with browsers via extensions for autofill. Such an architecture reduces single points of failure common in cloud password managers. It empowers users with full cryptographic control over their credentials.

32. Network Security & Data Engineering

A log analytics platform that correlates events from firewalls, web servers, IDS, and authentication systems. Logs are normalized, indexed, and visualized into unified attack graphs. Analysts can query by entity (e.g., user, IP, domain) to reconstruct intrusion campaigns. The system employs scalable big-data backends (e.g., Elasticsearch or ClickHouse). Alert rules flag multi-step attacks spanning multiple log sources. It bridges raw event streams into actionable cyber threat intelligence. Such a system strengthens security monitoring and forensic workflows.

33. Machine Learning & Operating Systems

An intelligent prefetching system that predicts file access patterns. Using ML models trained on historical process usage, it loads files into cache before applications request them. This reduces latency for frequently used workflows (e.g., IDEs or scientific tools). The system dynamically adapts as user habits evolve. Unlike traditional static prefetching, it captures contextual patterns like time-of-day or process chains. This integration improves system responsiveness without wasting I/O bandwidth. It brings predictive intelligence into OS-level caching.

34. Software Engineering & Data Engineering

An automated data quality validation framework for pipelines. It enforces schema rules, type checks, null constraints, and business-specific validation (e.g., email format). Failures are flagged before data propagates downstream. The framework integrates with CI/CD to ensure data quality is treated like software quality. Reports highlight anomalies, trends, and recurring issues. It can also auto-suggest cleansing operations. This minimizes pipeline corruption risks and enhances trust in analytics. It bridges software engineering rigor with data engineering workflows.

35. Cryptography & Network Management

A network-wide certificate authority (CA) that issues, renews, and revokes TLS certificates for devices. It distributes trust anchors securely across the network. Automation ensures certificates are rotated periodically without human intervention. Lightweight protocols reduce overhead for IoT or embedded systems. The CA integrates with identity management for role-based access. Revocation lists and OCSP are supported to invalidate compromised credentials. Such a system unifies cryptographic trust across heterogeneous devices.

36. Web Technologies & Network Security

A reverse proxy that doubles as a web application firewall (WAF). It inspects HTTP requests and blocks attacks like SQL injection, XSS, and CSRF. Rule-based and anomaly-based detection models provide layered defense. The proxy also handles SSL termination, caching, and load balancing. It integrates with dashboards for monitoring suspicious traffic. Administrators can define custom rules for application-specific threats. By combining proxying and WAF, it improves security without sacrificing performance.

37. Data Engineering & Operating Systems

A custom memory allocator with detailed logging capabilities. It tracks allocation sizes, frequency, and lifespan at runtime. Engineers can query logs to detect leaks, fragmentation, or performance bottlenecks. The allocator visualizes memory hot-spots and usage patterns over time. It can also simulate different allocation strategies for optimization. By merging OS-level memory management with data engineering insights, it aids debugging of complex applications. This tool is critical in systems programming and performance engineering.

38. Network Management & Machine Learning

A natural language processing (NLP) system for triaging network incident tickets. It categorizes issues, extracts key entities, and recommends probable fixes. The model learns from historical resolutions and administrator feedback. It reduces mean time to resolution by assisting operators with contextual insights. The system integrates with ITSM platforms like ServiceNow or Jira. It automates ticket handling while allowing human-in-the-loop verification. This improves efficiency in large-scale network operations centers.

39. Cryptography & Web Technologies

A hybrid encryption–based secure file-sharing platform. Files are encrypted with symmetric keys (AES) for efficiency, while keys are protected with recipient public keys (RSA/ECC). Users exchange encrypted keys securely via the platform. Files are shared without central servers, leveraging P2P or distributed storage. Metadata is minimized to avoid leakage of sender/receiver relationships. Client-side encryption ensures the service provider never sees plaintext. Such a system balances usability with strong cryptographic guarantees.

40. Operating Systems & Network Security

A process-aware firewall that controls outbound and inbound traffic at the application level. Unlike traditional firewalls, it enforces policies per process rather than per port/IP. It prevents malicious apps from secretly communicating over the network. The firewall integrates with OS-level process identifiers for accuracy. Administrators can define rules based on executable signatures. Logs show which processes attempted which connections. This fine-grained approach strengthens endpoint security.

41. Machine Learning & Data Engineering

An intelligent log classifier that automatically labels unstructured log data. Using NLP-based models, it groups entries by categories such as error type, component, or severity. This reduces manual log triage and speeds up root cause analysis. The system continuously adapts to new log patterns. It integrates into existing ELK (Elasticsearch-Logstash-Kibana) or Splunk stacks. By bridging ML and data pipelines, it enables smarter observability. It reduces noise and highlights actionable log insights.

42. Software Engineering & Cryptography

A secure random number generation library integrated into language runtimes. It wraps OS-provided entropy sources with cryptographic algorithms like ChaCha20 or AES-CTR DRBG. The library ensures forward and backward secrecy even if state is compromised. Developers can use it for key generation, nonces, or session IDs without worrying about weak entropy. It is packaged as a standard module to reduce reliance on unsafe custom RNGs. This improves cryptographic hygiene across ecosystems.

43. Network Security & Operating Systems

A custom honeypot that emulates a vulnerable system. Attackers are lured in and their behavior is logged in detail. It mimics OS-level responses to make the environment believable. Data collected includes malware samples, exploit patterns, and attacker tactics. It supports configurable profiles (e.g., web server, IoT device) for diverse threat research. Logs are streamed securely for analysis without risk of system compromise. Honeypots provide invaluable insights into attacker behavior and emerging threats.

44. Data Engineering & Web Technologies

A web-based dashboarding platform that automatically generates visualizations. It connects to SQL/NoSQL databases and infers schema types. For each dataset, it proposes default charts such as bar plots, time-series graphs, or heatmaps. Users can customize charts through drag-and-drop controls. The system supports live query updates and collaborative sharing. This reduces the barrier to analytics for non-technical users. It combines the accessibility of web apps with the depth of BI (Business Intelligence) tools.

45. Cryptography & Operating Systems

A Linux kernel module that enforces execution policies based on cryptographic signatures. Each executable must be signed by a trusted authority before being run. Unsigned or tampered binaries are blocked at load time. This prevents execution of unauthorized code, reducing malware risks. The policy integrates with system logs for auditing. It can also enforce role-based policies by mapping signatures to user groups. Such a mechanism aligns OS security with cryptographic trust anchors.

46. Network Management & Data Engineering

A network flow analyzer that monitors bandwidth usage in real time. It identifies “top talkers” and heavy bandwidth consumers. Flow data is stored in a time-series database (e.g., InfluxDB) for historical analysis. Administrators can query trends and set thresholds for alerts. Visual dashboards show usage distribution by device, port, or protocol. The system integrates with NetFlow/sFlow exporters. It helps optimize capacity planning and detect anomalies.

47. Web Technologies & Software Engineering

A UI framework for building component-based web apps. It enforces accessibility and responsiveness checks during development. Automated testing ensures color contrast, ARIA compliance, and mobile scaling. Developers receive real-time warnings if accessibility is broken. Components are optimized for reusability and modularity. The framework can plug into CI pipelines to guarantee UX quality. It merges modern software engineering with inclusive design principles.

48. Operating Systems & Machine Learning

A neural network–based CPU load predictor for energy optimization. It forecasts upcoming load and proactively adjusts CPU frequency or power states. Unlike reactive governors, it minimizes energy waste while avoiding performance drops. The model adapts to workload characteristics over time. It integrates with OS kernel power management frameworks. The system is especially beneficial in mobile and embedded devices. It introduces ML intelligence into energy-aware operating systems.

49. Data Engineering & Cryptography

A multi-party computation (MPC) framework enabling joint analysis of private data. Each party contributes encrypted shares of their dataset, and computations are done without revealing raw inputs. Supported functions include summations, averages, and ML model training. The framework is built for scalability, handling large datasets across distributed parties. Applications include finance, healthcare, and cross-institutional research. By merging MPC with data pipelines, it enforces privacy-preserving analytics.

50. Network Security & Machine Learning

A phishing detection system using ML classifiers trained on website data. It analyzes HTML structure, URL features, and content semantics. The system uses models like random forests or deep learning to distinguish phishing from legitimate sites. It adapts as attackers change tactics by continuous retraining. Browser plugins or network filters can integrate it for user protection. This reduces phishing-related credential theft. By automating detection, it complements traditional blacklists with proactive defense.

Cisco Confidential