
Building Production Multimodal AI Agents
Building Production Multimodal AI Agents is a technical book about engineering agents that work across images, video, screens, documents, audio, and text. Its focus goes beyond prototypes to retrieval, memory, evaluation, security, and deployment practices, making it most relevant to readers building real applications with Python, APIs, and basic LLM knowledge.
At a glance
- Covers workflows involving text, images, speech, video, and documents.
- Explores agent orchestration, tool use, state, memory, and retrieval.
- Discusses evaluation, security, observability, latency, cost, and deployment.
Reference price: $49.99
Price checked: Nov 20, 2026
View offerAhead of Black FridayExplore the selectionPros and cons
Pros
- Pairs multimodal input processing with practical production concerns such as evaluation, tracing, security, and failure recovery.
- Includes material on grounding outputs with retrieval and preserving evidence across interactions.
- Addresses screen agents and human review for workflows where actions need safeguards.
Technical specifications
9 specifications| Publisher | Packt Publishing |
|---|---|
| Publication date | November 20, 2026 |
| Edition | 1st |
| Language | English |
| Print length | 730 pages |
| ISBN-10 | 1809315468 |
| ISBN-13 | 978-1809315465 |
| Weight listed | 1.14 pounds |
|---|---|
| Dimensions | 7 x 0.52 x 10 inches |
What to know before buying
- The supplied listing gives a publication date of November 20, 2026, so confirm availability and edition details if timing matters.
Who is it for?
A good fit for
- AI and software engineers moving from text-only chatbots to multimodal applications.
- ML engineers, solution architects, and MLOps or LLMOps practitioners building agent systems.
- Readers comfortable with Python, APIs, and basic LLM concepts.
Less suited to
- Readers seeking an introductory guide that assumes no programming or LLM background.
- Readers looking only for theory without production-oriented engineering topics.
Frequently asked questions
What topics does the book cover?
The listing describes material on vision, video, screens, documents, audio, retrieval, memory, orchestration, evaluation, security, and deployment.
What background does the listing recommend?
It says readers should be comfortable with Python, APIs, and basic LLM concepts. Prior experience building multimodal agents is not required.
Is this focused only on building a prototype?
No. The described coverage includes production concerns such as evaluation, observability, security, latency and cost management, failure recovery, and deployment monitoring.























