AI Engineering: Building Multi-Modal Intelligent Systems with Vision, Language, and Audio From LLM Fine-Tuning to Voice Agents, AR Interfaces, and Real-World Deployment

Ara, Husn

/ ISBN 13: 9798296089038

Verlag: Independently published, 2025

Print on Demand. Bestandsnummer des Verkäufers I-9798296089038

Inhaltsangabe:

AI Engineering: Building Multi-Modal Intelligent Systems with Vision, Language, and Audio
From LLM Fine-Tuning to Voice Agents, AR Interfaces, and Real-World Deployment

Unlock the future of artificial intelligence with practical, production-ready multi-modal engineering.

This hands-on guide is built for developers, researchers, and AI professionals who want to go beyond chatbots and dive into building intelligent systems that understand text, images, audio, and human intent — all in one pipeline.

Whether you're fine-tuning large language models (LLMs) or creating voice-driven AR interfaces, this book walks you through the real engineering decisions, tools, and architectures needed to bring multi-modal AI to life.

What You'll Learn:

• Fine-tuning Large Language Models (LLMs): Train and adapt models like GPT-2, LLaMA, and Mistral for custom tasks using Hugging Face, LoRA, QLoRA, and PEFT.

• Voice Interfaces: Combine Whisper, LLMs, and Bark/Tortoise TTS to build interactive speech-driven assistants.

• Computer Vision + Language: Use models like BLIP, CLIP, and DETR to connect what systems see to what they say and understand.

• Instruction Tuning & Hyperparameter Optimization: Build smarter, domain-specific models with efficient training workflows.

• Multi-Modal Pipelines: Chain audio, image, and text inputs for question answering, summarization, tutoring, and AR/robotic control.

• Real-Time Interfaces: Deploy intelligent agents using FastAPI, Streamlit, Gradio, Docker, and Hugging Face Spaces.

• Edge & Offline Deployment: Optimize models with ONNX, quantization (4-bit, 8-bit), and TensorRT for low-latency inference on CPU/GPU.

Use Cases Covered:

• Smart document summarizers with OCR + TTS

• Voice-enabled image assistants

• Emotion-aware agents

• Virtual tutors

• AR-enhanced AI interfaces

• Robotic perception + control from voice/image input

• Secure, multilingual, and privacy-conscious AI systems

Tools & Frameworks Inside:

• Python, PyTorch, Hugging Face Transformers

• LangChain, OpenCV, Whisper, TTS, BLIP

• ROS, Unity (AR/VR), Gradio, Streamlit

• Docker, FastAPI, gRPC, TorchServe

Built for engineers. Written with depth. Designed for real-world impact.

If you're ready to build intelligent multi-modal agents that understand the world like humans do — across speech, vision, and language — this book gives you the complete roadmap.

Perfect for:
Machine learning engineers, data scientists, AI product developers, researchers, robotics engineers, and anyone building cutting-edge AI systems.

Bibliografische Details

Titel: AI Engineering: Building Multi-Modal ...
Verlag: Independently published
Erscheinungsdatum: 2025
Einband: Softcover
Zustand: New