NOW PLAYING

Écoute chérie

Vendredi sur Mer

➜~$ whoami

Harsha
Kondaveeti

AI/ML Engineer · Deep Learning · ML Systems

➜~$ cat about.txt

I build intelligent systems focused on Applied ML, Inference Optimization, and Deployment. My work seamlessly bridges the gap between cutting-edge research and low-latency production systems.

➜~$ ls -la Links/

resume.pdf linkedin.sh github.sh email.txt

➜~$Press Ctrl+M to interact

Get To Know Me

About Me

I'm an AI/ML enthusiast currently pursuing my B.Tech in Computer Science at IIIT Raichur, complemented by a Minor in Modern Machine Learning from IIIT Hyderabad.

My passion lies in building production-ready ML systems that bridge the gap between research and real-world applications. I specialize in developing RAG-based NLP pipelines, optimizing inference performance, and deploying scalable ML solutions.

Beyond coding, I've contributed as a Teaching Assistant for core CS courses, helping 200+ students strengthen their fundamentals. I'm also an active open-source contributor with 1.6K+ contributions across GitHub and GitLab.

IIIT Raichur

B.Tech in Computer Science

Aug 2023 - May 2027

IIIT Hyderabad

Minor in Modern Machine Learning

Aug 2025 - May 2026

0th

National SIH Rank

Students Mentored

OS Contributions

Career Path

Experience

AI Engineer Intern

OSCOWL AI

Remote

Feb 2026 - Present

Developing a real-time speech translation system integrating Speech-to-Text (STT), Neural Machine Translation (NMT), and Text-to-Speech (TTS) components. Building low-latency inference pipelines and deploying scalable microservices on AWS for production-ready multilingual speech processing.

STTNMTTTSAWSMicroservicesLow-Latency

AI Developer Intern

Viswam AI

Hyderabad, India

Jun 2025 - Aug 2025

Developed and optimized RAG-based NLP pipelines for Indic languages, focusing on retrieval relevance and contextual grounding. Built modular ingestion, embedding, and hybrid retrieval pipelines, and deployed Streamlit demos on Hugging Face Spaces.

RAGNLPHugging FaceStreamlitLLMs

Lead - Website Development Team

IIIT Raichur

Raichur, India

Oct 2024 - Present

Architected modular CMS workflows with CI/CD automation and CRM integrations for the institute website. Led Cloudflare CDN, caching, and autoscaling deployment, reducing manual update effort by 48+ hours per cycle.

CI/CDCloudflareCMSDevOps

Selected Work

Featured Projects

A showcase of my recent academic, research, and personal projects spanning machine learning to modern web development.

Peptide Ion Fragmentation Prediction using GNNs

Developed Graph Neural Network models for predicting peptide fragment ion probabilities from DIA mass spectrometry data, modeling peptide-fragment relationships to improve identification accuracy and cross-instrument generalization.

PyTorchGNNsMass SpecDeep Learning

View Repository

Re-Search - Literature Intelligence

Designed a controlled multi-agent research workflow incorporating task decomposition, verification and critique loops, to improve reliability and reduce hallucination risk in automated literature analysis.

LangGraphMulti-Agent SystemRAGLLMs

View Repository

Hybrid-DIP-ViT Underwater Image Segmentation

Implemented and evaluated a hybrid DIP + Vision Transformer segmentation pipeline, achieving ~6% mIoU improvement over U-Net baselines on low-visibility underwater datasets while maintaining stable convergence under noise.

Vision TransformersSegmentationU-NetCV

View Repository