Default profile banner
PH

Prathap HM

@user.2533760

AI/ML Developer at EMotorad

Bengaluru, India

https://linkedin.com/in/prathaphm

EMotoradMS Ramaiah University of Applied Sciences

Prathap HM is an AI/ML Developer at EMotorad and a top-ranked B.Tech graduate from MS Ramaiah University of Applied Sciences with a CGPA of 9.66 and the university's gold medal. He specializes in conversational AI, RAG systems, NLP, deep learning, and model optimization. His work includes developing a conversational AI system with a 500,000-entry FAISS vector database and a voice-enabled interaction system with sub-1.5-second latency, as well as data analytics roles at Toyota Kirloskar Motor.

Experience

AI/ML Developer

EMotorad

Full-time•Aug 2024 - Present•Bengaluru, India

Foundational member of AI/ML Team. Designed and developed a conversational AI system integrating a LLM based on Transformer Architecture, reducing response inconsistencies by 40%. Implemented RAG with a 500,000-entry FAISS vector database, reducing factual inconsistencies by 30% with retrieval latency under 300ms. Developed voice-enabled interaction system with STT and TTS with average latency under 1.5 sec. Applied model optimization including quantization (INT8), pruning, and knowledge distillation, reducing model size by 50% while maintaining 95% accuracy.

Data Analyst

Toyota Kirloskar Motor

Full-time•Dec 2023 - Jun 2024

Developed and implemented a data-driven approach to optimize production line scheduling, resulting in 20% reduction in production downtime and 15% increase in overall output. Developed automated predictive maintenance model using SQL, Python and Power BI, reducing unplanned downtime by 25%. Presented data-driven insights to senior management.

Education

MS Ramaiah University of Applied Sciences

Bachelor of Technology

Jan 2018 - Jan 2022•Grade: 9.66/10 CGPA

Skills

Python
Mathematics
Statistics
Machine Learning
Deep Learning
TensorFlow
Pytorch
Computer Vision
Time Series Prediction
Transformers
NLP
Generative AI
Retrieval Augmented Generation
Git
SQL
Power BI
Model Optimization
Quantization
Pruning
Knowledge Distillation
FAISS
Speech-to-Text
Text-to-Speech