Open to opportunities

Saketh K

@sakethk

Message

I build edge AI systems for speech, radar sensing, and real-time computer vision.

India

Message

What I'm looking for

I'm seeking roles in embedded AI, edge computing, radar systems, or speech ML where I can build efficient real-world systems balancing accuracy, latency, compute efficiency, and hardware integration.

At IIIT Hyderabad, I built speech ML preprocessing pipelines for 500+ hours of audio and improved baseline classification accuracy by approximately 15% through model benchmarking and error analysis. I also developed automated labeling and validation tooling for large-scale speech corpora while weighing latency and accuracy for real-time deployment.

I've built embedded and signal-processing systems spanning ESP32 RFID access control, radar-based object classification on Raspberry Pi, and UAV magnetic anomaly sensing. My work has earned a national Smart India Hackathon win, a published patent for rail fault monitoring, and an accepted ICTCS 2025 paper on real-time deepfake detection.

Experience

Work history, roles, and key accomplishments

Research Intern — Speech ML & Audio Processing

IIIT Hyderabad

Jun 2025 - Mar 2026 (9 months)

Engineered end-to-end audio preprocessing pipelines for 500+ hours of speech data, including segmentation, normalization, and feature extraction. Developed and benchmarked speech classification models, contributing to a ~15% improvement in baseline accuracy.

Python Librosa Spectrograms Speech Processing Audio Processing Machine Learning Benchmarking