Building AI systems from research to production.

I'm Aditya Mhaske. 5 IEEE publications, 2 patents, and currently engineering multi-tenant RAG infrastructure at enterprise scale.

Aditya Mhaske

Selected Work

Production AI systems with measurable impact.

IndusAI — Industrial RAG Platform
RAGLLM SystemsProduction AI

IndusAI — Industrial RAG Platform

Privacy-first industrial AI assistant analyzing PLC fault logs with hybrid RAG retrieval, local LLM reasoning (Ollama + Mistral 7B), deterministic statistics, and strict hallucination guarding. Runs 100% locally with multi-tenant vector search via Qdrant.

100+ environments>99% redaction30x latency reduction
View on GitHub
Multimodal Depression Analysis
Deep LearningNLPComputer Vision

Multimodal Depression Analysis

Real-time system recognizing depression by extracting features from facial expressions, EEG signals, and speech modules using multimodal deep learning. Supported by peer-reviewed research publications.

3 modalities91% precisionIEEE published
View on GitHub
Object Detection & Localization
Computer VisionAutonomous Systems

Object Detection & Localization

YOLOv3 trained on the COCO dataset for real-time object detection and localization on road scenes. Related to autonomous vehicle patents and deployed in a braking system prototype.

Real-time detection2 patents filedYOLOv3 + COCO
View on GitHub

Technical Toolkit

Modern infrastructure for advanced AI development.

ML & AI

Scikit-learnXGBoostTensorFlowPyTorchTransformersLLMsRAG

Languages

PythonSQLJavaC++R

Data & Ops

AirflowAWSSnowflakeDockerCI/CDETL

Core Pillars

Deeply technical, focused on scalability and real-world impact.

🔬

Research → Production

  • Built NLP pipeline processing 50K+ PLC logs daily.
  • Hybrid search with 20% Recall@5 improvement.
  • End-to-end CI/CD for AI microservices.
📊

Statistical Intelligence

  • A/B testing frameworks driving 17% engagement lifts.
  • Cross-classified multilevel models.
  • Behavioral analysis on 100M+ row datasets.
🤖

LLM & AI Systems

  • Retrieval-augmented generation with hallucination guarding.
  • Multi-tenant vector search across 100+ environments.
  • >99% redaction accuracy.

What Others Say

Recommendations from faculty and research mentors.

Aditya Mhaske is a very bright, extremely skilled, and hardworking MS in Data Science. Aditya's analytic, coding, programming, and presentation skills place him in the top 1% of data science students I have interacted with at Indiana University over the past 12 years. Without hesitation, I recommend Aditya to any employer seeking to hire him.

Prof. Lopo Rego

Prof. Lopo Rego

Chairperson, Marketing Dept · Kelley School of Business, IU

Aditya is incredibly hard-working, enthusiastic, and proactive. He has significantly helped me forward my research on television advertising, social media, and political marketing. I would be more than happy to elaborate on my recommendation for Aditya.

Prof. Beth L. Fossen

Prof. Beth L. Fossen

Assistant Professor of Marketing · Kelley School of Business, IU

Aditya proved to be a hard-working and ambitious individual always eager to learn. He developed projects in deep learning and signal processing that showcased his technical expertise. His in-depth analysis and forward-thinking mindset helped propose improvements for potential problems.

Prof. Shamla Mantri

Prof. Shamla Mantri

Associate Professor · MIT World Peace University

Let's build something extraordinary.

Open to engineering roles and research collaborations in ML, LLM systems, and enterprise AI.