Krishnan Raghavan

MSc in Artificial Intelligence and Data Science (Keele University, UK) and BE in Computer Science and Engineering (College of Engineering, Guindy, India). Experienced in applying machine learning and deep learning techniques to solve real-world problems, with project work spanning phishing detection, music recommendation, face recognition, and demand forecasting. Demonstrated ability to design, test, and deploy AI-driven systems with measurable outcomes, including achieving >95% phishing recall and 97% recommendation accuracy.

Seeking entry level roles in software engineering or data / artificial intelligence roles.

Software Engineer (Academic Project) (2025-05 – 2025-09)

Rising phishing attacks compromise security. Built PhishBERT, a Machine Learning model achieving >95% recall.

• Developed PhishBERT, a hybrid natural language processing model in PyTorch trained on 30,000+ emails. Built a Flask-based User Interface for real-time testing and suitable for non-technical users.
• Achieved >95% phishing recall, delivering a deployable solution with real-world applicability in cybersecurity.

Software Engineer (Academic Project) (2023-12 – 2024-05)

Traditional recommendation engines fail to capture nuanced user behavior and emotional states, leading to limited personalization.

• Designed a hybrid recommender integrating genre classification and emotion recognition. Implemented models in TensorFlow and optimized with Structured Query Language + Pandas for large datasets.
• Delivered 97% accuracy in recommendations and demonstrated integration into large-scale platforms.

Data Scientist (Academic Project) (2025-01 – 2025-03)

Ride-sharing companies struggle with unpredictable demand influenced by various factors.

• Processed and analyzed 40,000+ data points using Python and SQL. Built predictive models and generated trend visualizations.
• Produced actionable insights enabling ride-share companies to optimize fleet allocation and reduce downtime.

Software Engineer (Academic Project) (2022-09 – 2022-12)

Real-time face recognition on constrained hardware was inefficient and error-prone.

• Built a Python-based application with OpenCV and Haar cascades; tuned algorithms for improved recognition accuracy.
• Optimized the system and achieved real-time performance with high reliability on end-point devices with critically low resources.

Software Engineer (Academic Project) (2022-03 – 2022-06)

Need for robust recognition of handwritten digits for digitization and automation tasks.

• Trained a model on benchmark datasets; applied regularization to improve generalization. Conducted multiple test cases.
• Achieved >98% classification accuracy, validating system robustness across varied handwriting styles.

MSc in Artificial Intelligence and Data Science – Keele University, UK (2024-09 – 2025-09)

BE in Computer Science and Engineering – College of Engineering, Guindy, India (2019-12 – 2024-05)