About
Applied AI/ML Engineer building agentic AI systems, production LLM pipelines, and scalable cloud deployments. Architected enterprise RAG platform serving 230+ users at 75% lower cost; shipped multi-agent LangGraph research assistant with corrective RAG and hallucination detection. Published researcher (Springer, IEEE) in deep learning, NLP, and computer vision.
Experience
MiHIN
AI Research Intern
Lansing, MI
Jul 2025 - Present
- Architected production Document AI/RAG platform on AWS Bedrock, OpenSearch Serverless (k-NN vector search), and Lambda, enabling 230+ employees to query 1T+ documents via natural language at 75% less than competitors.
- Engineered NLP ingestion pipeline with document parsing, Titan V2 embedding generation (1,024-dim), configurable chunking, vector indexing, and retrieval; incorporated human-in-the-loop (HITL) feedback to tune chunk size, overlap, and top-k parameters, reducing LLM hallucination rates by 30%.
- Developed AI governance framework aligned with HITRUST CSF, NIST, HIPAA; spearheaded AI Task Force training across DevOps, Legal, and HR, authoring playbooks that enabled 5+ cross-functional teams to integrate generative AI into production workflows.
- Delivered serverless REST API with API Gateway, Lambda, Cognito (SSO + 2FA) supporting 230+ users; presented cost analysis to CEO, securing approval for org-wide pilot cutting costs by 75%.
- Reduced mean time to detection by 60% through CloudWatch dashboards, structured logging, and custom alarms for real-time monitoring; implemented error handling and retry logic across Lambda functions with dead letter routing for fault tolerance.
Rygen BioPharma
Data & Software Development Analyst
Hyderabad, India
Mar 2023 - Aug 2024
TreoSoft IT Solutions
Data Science Intern
Bengaluru, India
Jun 2022 - Sep 2022
Projects
ScholarAgent
Multi-Agent Research Assistant
Multi-agent research assistant with a 7-node LangGraph pipeline featuring corrective RAG, query rewriting, and hallucination detection.
- Orchestrated multi-agent research assistant with 7-node LangGraph pipeline (router, retriever, grader, rewriter, generat...
- Integrated ChromaDB vector search with HuggingFace sentence-transformer embeddings indexing 10K+ arXiv/PubMed papers; co...
MCP Healthcare Server
Model Context Protocol Server
MCP server exposing 6 healthcare tools via FastMCP, integrated with 4 federal APIs for drug safety, clinical trials, ICD-10 lookup, and more.
- Launched Model Context Protocol server exposing 6 healthcare tools (drug safety, clinical trials, ICD-10 lookup, medical...
LinkVault
Go REST API Platform
Go REST API platform with JWT and API-key authentication, Redis-backed rate limiting, and PostgreSQL full-text search.
- Engineered Go REST API platform with JWT and API-key authentication, Redis-backed rate limiting, and PostgreSQL full-tex...
Distributed Task Queue
Go + gRPC + Kubernetes
Distributed task queue in Go with gRPC/Protobuf service interfaces, Redis-backed priority queue, and Kubernetes deployment with autoscaling.
- Built distributed task queue in Go with gRPC/Protobuf service interfaces, Redis-backed priority queue (sorted sets), and...
- Deployed on Kubernetes with Helm charts and horizontal pod autoscaling on custom Prometheus metrics (queue depth); instr...
Neural Network Training Dynamics & Pruning
PyTorch Research Project
Research into CNN vs. Transformer representation learning via UMAP/t-SNE and CKA similarity, with L1 unstructured pruning analysis.
- Trained ResNet-18 and ViT-Tiny on CIFAR-100 using transfer learning; analyzed CNN vs. Transformer representation learnin...
LLM-Driven Procedural Content Generation
LLMs4PCG - AAAI 2026
Applied open-source LLMs to procedural game-level generation with constraint satisfaction and ViT-based validation.
- Applied latest open-source LLMs (Phi-3, Gemma-2, Qwen-2.5) to procedural game-level generation; designed prompt compilat...
AI ChatBot
Multi-Provider LLM Service
Multi-provider generative AI web service with FastAPI backend and LangChain orchestration, featuring provider failover for high availability.
- Deployed multi-provider generative AI web service with FastAPI backend and LangChain orchestration; implemented async AP...
Skills
Publications
Potato Disease Classification Using Diverse Feature Extraction Methods and Machine Learning Models
C. Shanmukha Srinivas Sai et al.
HIS 2023 (Hybrid Intelligent Systems), 2023
Tea Leaf Disease Classification with Ensemble Stacking
Somesh K, C. Shanmukha Srinivas Sai et al.
ICDLAIR 2024 (International Conference on Deep Learning, Artificial Intelligence and Robotics), 2024
Education
North Carolina State University
MS Computer Science (Data Science Track)
GPA: 3.71 | Aug 2024 - Jun 2026
National Institute of Technology Puducherry
BTech Computer Science & Engineering
GPA: 8.48/10 | Dec 2020 - Mar 2024
Leadership & Achievements
President
Rotaract Club, NIT Puducherry
2022 - 2023
Organized 10+ community service events including blood donation camps and social awareness campaigns.
Vice-President
ACE (CS Association), NIT Puducherry
2022 - 2023
Led workshops, hackathons, and industry talks for the computer science student body.
Co-founder
Zer01Coded, NIT Puducherry
2021 - 2022
Founded coding club and conducted Python programming classes for 150+ students.
Achievements
Beyond the Code
Basketball
Click to flip
Played competitively and won tournaments in college.
Astronomy
Click to flip
Love stargazing and keeping up with space exploration missions.
Cyberpunk 2077
Click to flip
A game that keeps pulling me back into Night City.
Origami
Click to flip
Folding paper into intricate designs is surprisingly meditative.
Weightlifting
Click to flip
Consistent gym-goer focused on strength training.
Teaching
Click to flip
Taught Python to 150+ students through Zer01Coded club.
Contact
{
"email": "schatad@ncsu.edu",
"linkedin": "linkedin.com/in/shanmukha-chatadi",
"github": "github.com/shinegami-2002",
"location": "Raleigh, NC (Open to Relocation)",
"status": "Open to opportunities"
}