Dev Savaj

About Dev Savaj

I am a Generative AI Engineer focused on transforming Large Language Models (LLMs) from experimental prototypes into scalable, enterprise-grade AI systems. Currently at Ascendion, I work at the intersection of AI research and real-world product engineering, bridging the gap between impressive AI demos and business-critical solutions. My mission is simple: Move AI beyond conversation. Build systems that reason, automate, and deliver measurable impact. I specialize in designing intelligent platforms where AI is not just a feature — but a core decision-making layer powering modern applications. What I Build & Solve Anyone can build a chatbot. Engineering reliable AI systems that operate securely at enterprise scale is the real challenge. My core focus areas include: Advanced RAG Architectures Designing high-accuracy Retrieval-Augmented Generation pipelines using semantic search, embeddings, and vector databases to minimize hallucinations and maximize grounded responses. LLM Orchestration & AI Agents Building multi-step reasoning systems and autonomous agents using frameworks like LangChain, LlamaIndex, CrewAI, and AutoGen. Prompt Engineering & Evaluation Moving beyond prompts into structured reasoning, few-shot optimization, automated evaluation, and response quality benchmarking. Production AI Engineering Optimizing latency, inference cost, observability, security, and responsible deployment of AI inside enterprise ecosystems. 🧠 Technical Toolkit AI / Machine Learning Generative AI & Large Language Models (GPT-4, Claude, Llama) NLP, Embeddings, Fine-Tuning, Agentic AI Systems Retrieval Systems & Knowledge Grounding Frameworks & Infrastructure LangChain • LlamaIndex • CrewAI • AutoGen Vector Databases: Pinecone, Weaviate, Milvus RAG Pipelines & AI Workflow Orchestration Engineering & Architecture Scalable AI Pipelines API & Tool Integration Secure AI System Design Performance Monitoring & Evaluation My Perspective on AI Generative AI evolves at a weeks-not-months pace. The engineers who will shape the future are not those who only experiment with models — but those who productionize intelligence. I focus on building AI systems that are: Reliable Explainable Scalable Business-driven 🤝 Let’s Connect I’m always open to conversations around: Enterprise AI Architecture Intelligent Agent Systems Applied Generative AI Research AI Product Engineering Opportunities If you\'re turning AI hype into real utility, we’ll get along well.

Experience

1. Generative Ai Engineer

   Ascendion

   Feb 2026 — Present · IN

   I’m thrilled to announce that I’ve joined Ascendion as a Generative AI Engineer Intern! I am eager to contribute to the next generation of Agentic AI and help build high-impact, enterprise-grade solutions. My focus will be on:Architecting Scalable AI: Implementing robust Generative frameworks and RAG architectures.Driving Engineering Excellence: Enhancing software delivery through autonomous agentic workflows.

Education

• Veer Narmad South Gujarat University, Surat

  Bachelor of Computer Applications, Computer Science

• Uka Tarsadia University

  Master of Computer Applications - MCA, Computer Science

• Veer Narmad South Gujarat University, Surat

  BCA, Computer Science

  2022 — 2024

• UKA TARSADIA UNIVERSITY

  Master of Computer Applications - MCA, Mathematics and Computer Science