23/10/2026SHIVANSHU KUMAR ANAND
SHIVANSHU KUMAR ANAND
AI Solutions Lead / Associate AI Architect (GenAI/Agentic AI)
+91-9810713113shivanshu.anand27@gmail.comlinkedin.com/in/shivanshu-anand-5b36b171/
Professional Summary
Visionary AI Solutions Architect with 10+ years of hands-on experience in architecting, governing, and scaling AI delivery practices across Generative AI, Agentic AI, and applied ML engagements. Deep expertise in end-to-end GenAI solution architecture, LLM workflow orchestrations (LangGraph, LangChain), and deploying multimodal agentic systems and SLM-based deployments.
Professional Experience
Senior Engineer (Lead)May 2022 - July 2026
Apexon • Bangalore, India
- Architected and delivered end-to-end Generative AI solutions, building self-corrective RAG systems and semantic search engines using Elasticsearch.
- Established MLOps principles, created scalable FastAPI microservices, and drove frontier vs. SLM model selection trade-offs.
- Managed evaluation frameworks, red-teaming protocols, and guardrails for hallucination control.
- Led a 30-member full-stack engineering team.
Senior Software DeveloperAug 2021 - Feb 2022
Times Internet Ltd.
- Implemented ML-based user suggestions, social feed ranking, and NLP content moderation.
Lead Frontend DeveloperOct 2020 - Jul 2021
MikeLegal Services
- Contributed to DocuSieve 1.0 architecture, incorporating NLP-based search, OCR text extractions, and document classification pipelines.
Co-founder & Frontend EngineerFeb 2015 - Mar 2020
Vaidik Technologies
- Designed AI recommendation systems for personalized learning and deployed OpenCV facial recognition logins.
Core Skills
Frontend ArchitectReact.jsNext.jsAgentic SystemsMulti-Agent OrchestrationLangGraph / LangChainGenerative AI & SLMsFine-Tuning (SFT, LoRA)RAG / Agentic RAGEval & GuardrailsElasticsearchKnowledge GraphsMLOps / GenAIOpsPython / FastAPITypeScript
Education
M.Tech in Advanced Networks
specialized in AI
IIIT Gwalior • 2017 - 2019
B.Tech in Computer Science
Graphic Era Deemed University • 2012 - 2016
Research Publications (4)
EXPLORE ALL →- A Multimodal EEG Framework for Cognitive Stress Detection with Retrieval-Augmented ExplainabilityCNN, Bi-LSTM & Sentence-BERT (99.31% EEGMAT Acc, 89.8% Expert Agreement)
- A Deep Learning Architecture with Quantum-Inspired Feature Fusion for Robust EEG-Based Drowsiness DetectionTemporal + STFT Spectral branches with Amplitude–Phase projection (88.10% Acc, 0.9197 ROC AUC)
- EEG-Based Classical Framework for Autism Spectrum Disorder DiagnosisContinuous Wavelet Transform (CWT) & CNN-ASD architecture (94.20% Acc, IEEE Xplore Record #11280701)
- Classification of Hematological Malignancies Using Peripheral Blood Smear ImagesConditional GAN data augmentation & deep CNN classifier (97.10% synthetic quality, IEEE Xplore Record #11295574)