AI & ML Engineer specializing in LLM Alignment (RLHF), Audio AI Workflows, and Multimedia Production.
Combining academic rigor at Aditya University with global commercial execution — including prompt evaluation, Labelbox speech annotation, functional web testing, and US-grade voiceover production for Babel Audio.
I am currently in my final year pursuing a B.Tech in Artificial Intelligence and Machine Learning at Aditya University, Surampalem (Graduating Class of 2027).
My experience crosses the boundary between machine algorithms and real-world human judgment. Over the last two years, I have trained enterprise generative models through RLHF and SFT evaluations, annotated complex audio streams with high phonetic precision using Labelbox, recorded broadcast voice assets for US clients, and tested web and mobile systems for critical user friction.
Aditya University, Surampalem, Andhra Pradesh
Key Coursework: Natural Language Processing, Deep Learning, Automatic Speech Recognition (ASR), Data Structures, Software Testing & QA, Applied Statistics.
Specialized in RLHF preference ranking, hallucination mitigation, safety compliance, and multi-turn instruction fine-tuning for frontier models.
Acoustic segmentation, speaker diarization, and phonetic transcription on Labelbox workflows for Automatic Speech Recognition pipelines.
Broadcast-standard voiceover recording for US-based Babel Audio, dynamic range mastering, noise suppression, and Premiere Pro video editing.
A blend of machine learning theory, data engineering craftsmanship, and multimedia production tools.
Prompt optimization, fine-tuning data audits, and model alignment.
Preparing high-fidelity datasets for deep learning and ASR models.
Auditing digital platforms to eliminate defects and user friction.
Enterprise studio voice artistry for commercial, narrative, and e-learning media.
Post-production, pacing, timeline synchronization, and audio-video integration.
Core languages and development frameworks for data and model pipelines.
Direct contributions across enterprise data operations, US-based audio narration, and quality assurance.
Generative AI Human Feedback & Evaluation
Labelbox Annotation Platform
Babel Audio (United States)
Web & Mobile Application Quality Verification
Real solutions bridging machine learning models, automated data workflows, and user interfaces.
A speech-to-text inference pipeline fine-tuning Whisper architecture on multi-accented speech samples. Incorporates automated spectral cleaning, reducing Word Error Rate on noisy background recordings.
Semantic retrieval engine designed to query dense technical PDFs and documentation. Implements dynamic chunking, vector indexing, and source citation to strictly prevent model hallucinations.
A light inspection dashboard built for audio annotators. Inspects waveforms, calculates SNR (Signal-to-Noise Ratio), verifies sample rates, and ensures audio meets Labelbox ingestion specs.
Interactive studio player showcasing acoustic range, broadcast pacing, and US commercial standards delivered for Babel Audio.
Produced by Satya Vardhan Pathagolusu
I am actively exploring Full-time Engineering Roles for the 2027 batch, high-impact LLM Training / Data Alignment contracts, and voiceover projects. Let's connect!
Leave a brief note and I will get back to you promptly.