Open to opportunities

Satyam Shukla

@satyamshukla1

Message

I build AI and machine learning systems for document intelligence, speech analytics, and multimodal prediction.

India

Message

What I'm looking for

I'm looking to build practical AI and machine learning products, especially LLM, RAG, multimodal, speech, and data-driven systems where I can improve performance at scale.

I'm building LLM-based systems at ICA Impetus, where I automated transcription, sentiment analysis, speaker diarization, and topic extraction for processed calls. I redesigned the platform with asynchronous, multithreaded processing, cutting transcription time by over 80% and supporting 10× higher concurrency with no performance degradation.

I've also built a RAG-based document assistant that processes 1,000+ page documents in under 30 seconds and achieves 87.82% accuracy, as well as Agroculture, a multilingual agricultural assistant with real-time weather, market-price, voice, and semantic-search capabilities.

My research at LNMIIT benchmarked 22 preprocessing algorithms across nine EEG/ECG datasets, showing preprocessing-driven performance differences of up to 15–25%.

Experience

Work history, roles, and key accomplishments

Current

AI/ML Engineer Intern

Current

ICA Impetus

Dec 2025 - Present (8 months)

Built an LLM-based transcription and analytics pipeline automating sentiment analysis, speaker diarization, and topic extraction, and engineered a multithreaded parallel processing pipeline reducing transcription time by over 80%.

LLM Python Multithreading Asynchronous Programming Sentiment Analysis Speaker Diarization Topic Extraction

Biomedical Data Analyst

LNMIIT

Aug 2025 - Nov 2025 (3 months)

Benchmarked 22 algorithms across 9 EEG/ECG datasets and designed 22 preprocessing pipelines using ASR, Wavelet Denoising, and ICA, observing performance variations of up to 15-25%.

Machine Learning Deep Learning Signal Processing EEG ECG Artifact Subspace Reconstruction ICA Python