AI Systems Design: RAG Pipelines and LLM Architecture | Board Infinity
Home/Courses/Software Development/AI Systems Design
Intermediate Course 1 of 3 · Managing AI Systems

AI Systems Design: RAG Pipelines and LLM Architecture

Design enterprise AI that holds up on technical and business feasibility - frame the problem, architect LLMs, vector databases and orchestration, then build a working RAG prototype in Streamlit.

Enroll on Coursera
2 weeks at 10 hrs/week 4 modules · 15 assignments Intermediate level Taught in English
Want a coach alongside the course? See 1:1 coaching
Board Infinity287 courses · 479,274 learners on Coursera
Loading video… Introduction to the course · 8:05

Problem to prototype

From a business need to a deployed RAG MVP

Certificate included

Shareable on LinkedIn the day you finish

For engineers and PMs

Technical depth with a business lens

6 Lakh+learners on our Coursera courses
10+domains covered
270+courses
57+specializations
01 About the course

From business problem to working prototype.

Four modules that design, build and refine an enterprise RAG system - and make the business case for it.

4
Modules
Scope, stack, data, prototype
36
Video lessons
Short and focused
15
Assignments
Quizzes and hands-on tasks
2 wks
At 10 hours a week
Self-paced, flexible deadlines

AI Systems Design teaches you to design and prototype enterprise-grade AI solutions that create real business value.

You learn to distinguish when to use predictive ML versus generative AI, align AI initiatives with business outcomes, and define success criteria that balance accuracy, latency, safety and cost. Then you diagram modern AI architectures and evaluate build-versus-buy decisions for models, vector databases and orchestration layers.

Hands-on, you implement ingestion, chunking and embedding strategies, compare RAG with fine-tuning, and finish by building an end-to-end RAG prototype with a Streamlit UI - tested and refined from its logs.

Inside the course
01Problem Formulation & AI Solution Scoping≈ 4h 02The Modern AI Stack≈ 5h 03Data Strategy for RAG and Fine-Tuning≈ 4h 04Prototyping & Building the Company Brain≈ 4h
02 Who it's for

Built for the people who decide how AI gets built.

Basic familiarity with AI/ML concepts helps but isn't mandatory - the course starts with foundational frameworks.

Engineering

Engineers

Designing and building AI systems that have to work in an enterprise, not just a notebook.

Product

Product managers

Framing AI problems around business metrics, and knowing when AI is the wrong answer.

Delivery

Technical program managers

Scoping AI initiatives, managing constraints and setting acceptance criteria.

Governance

AI governance officers

Understanding how AI systems are built, tested and monitored before you govern them.

03 What you'll learn

Four capabilities, in the order they build.

These are the four learning outcomes Coursera lists for the course - one for each module.

Module 1

Classify AI problems

Apply decision frameworks to classify AI problems and evaluate build-versus-buy trade-offs.

Module 2

Build end-to-end RAG pipelines

Use LangChain, ChromaDB and embedding models to retrieve context and generate answers.

Module 3

Design data strategies

Chunking, embedding selection and retrieval optimization for messy enterprise data.

Module 4

Prototype with Streamlit

Create interactive AI prototypes for real-world business applications.

1:1 coaching · optional add-on

Build with 1:1 coaching from industry experts.

A dedicated coach who designs AI systems for a living - someone who can pressure-test your architecture and your business case before anyone else does.

Book a demo session
  • A coach matched to your goal in 24 hours
  • A written roadmap you keep
  • Weekly sessions for as long as you want them

60-minute demo session · 100% refundable · no obligation to continue

Pressure-test your design

Architecture and build-vs-buy choices reviewed by someone who has shipped them.

Sharpen the business case

KPIs, ROI and acceptance criteria that stakeholders will sign off.

Improve your prototype

Retrieval quality, latency and grounding, debugged together.

Aim it at a role

Architect, AI PM or TPM - which path fits, and what to learn next.

2,000+Industry coaches
16Domains covered
4.8/5Average coach rating
48,000+Sessions delivered
04 Syllabus

Four modules, from problem to prototype.

Coursera lists this course at 2 weeks of study at 10 hours a week, fully self-paced. Every module lists its learning objectives, its content, and what you'll walk out with.

36 videos 12 readings 15 assignments 2 weeks at 10 hrs/week
Learning objectives - by the end of this module you can:
  • Choose between predictive ML and generative AI with a decision framework
  • Define KPIs, ROI and acceptance criteria for an AI initiative
  • Translate business needs into AI requirements within real constraints
  • Plan testing for non-deterministic outputs and explain AI behavior to stakeholders
What's inside
9 videos54 min
3 readings90 min
3 assignments + 1 plugin95 min
Topics covered
Career scope in AI systems Predictive ML vs generative AI KPIs, ROI & acceptance criteria Constraints & feasibility Non-deterministic outputs & UX Testing probabilistic systems
You'll leave with: a scoped AI problem with clear success criteria - and a case for where AI fits and where it doesn't.
Learning objectives - by the end of this module you can:
  • Identify the components of the AI stack: LLMs, vector stores and orchestrators
  • Trace the data flow from ingestion to embedding, retrieval and generation
  • Operate ChromaDB and FAISS with chunking and retrieval optimization
  • Apply chains, tools and agent patterns, and implement a simple RAG chain
What's inside
9 videos77 min
3 readings90 min
4 assignments105 min
Topics covered
Components of the AI stack Ingestion → retrieval → generation Embeddings & model types ChromaDB & FAISS Chains, tools & agents A simple RAG chain
You'll build: a first working RAG chain, on a reference architecture for enterprise RAG.
Learning objectives - by the end of this module you can:
  • Set chunk size, overlap and structure, and choose an embedding model
  • Measure retrieval accuracy
  • Compare RAG and fine-tuning on cost, latency and data sensitivity
  • Clean, normalize and tag enterprise data, and manage noise, PII and gaps
What's inside
9 videos61 min
3 readings90 min
4 assignments105 min
Topics covered
Chunking heuristics Embedding model selection Measuring retrieval accuracy RAG vs fine-tuning Cleaning & metadata tagging Data readiness risks
You'll leave with: a data strategy for messy enterprise data - and the evidence to choose RAG or fine-tuning.
Learning objectives - by the end of this module you can:
  • Implement ingestion, chunking and embedding, and validate retrieval quality
  • Build a Streamlit UI and integrate the retrieval chain into it
  • Deploy the MVP
  • Find hallucinations, fix latency and improve grounding from test logs
What's inside
9 videos59 min
3 readings90 min
4 assignments105 min
Topics covered
End-to-end RAG system Retrieval quality validation Streamlit for AI prototypes Deploying the MVP Hallucinations & irrelevant retrievals Latency & grounding
You'll build: a deployed "company brain" - an end-to-end RAG prototype with a Streamlit UI.
05 Tools & skills

The stack you design with, layer by layer.

The components of a modern AI system, and the skills you leave with.

Models

LLMs & embedding models

Choosing the paradigm and the model, from predictive ML to generative AI - Modules 1 to 3.

Vector databases

ChromaDB & FAISS

Storing and searching embeddings, with chunking and retrieval optimization - Module 2.

Orchestration

LangChain & LlamaIndex

Chains, tools and agent patterns, and when to use each - Module 2.

Prototype UI

Streamlit

The interactive front end for your RAG prototype, through to the deployed MVP - Module 4.

Skills you leave with

Twelve skills from the course, grouped the way AI systems work actually splits - designing the system, getting the data right, and building and refining it.

Architecture

How the pieces of an AI system fit together.

  • Systems architecture
  • Solution architecture
  • AI orchestration
  • LLM application

Data & retrieval

Turning messy enterprise data into the right context.

  • Retrieval-augmented generation
  • Embeddings
  • Vector databases
  • Unstructured data

Build & refine

From a prototype to something stakeholders can use.

  • Fine-tuning
  • Data preprocessing
  • User interface (UI)
  • Generative AI
Certificate

Proof you can link to.

Finish the graded work and Board Infinity issues a shareable certificate through Coursera - add it to your LinkedIn profile, resume or CV.

Issued on completion of the assignments and quizzes in all four modules. It also counts toward the Managing AI Systems specialization certificate.

On your profile in one click

Filed under Licenses & Certifications, where recruiters actually look.

Share it anywhere

On social media and in your next performance review.

Earned, not attended

Awarded on 15 assignments, including hands-on tasks - it certifies what you can design.

Two names on it

Board Infinity as the provider, Coursera as the platform.

06 Career impact

Every company wants AI. Few know how to scope, build and ship it.

These are the roles that rely on exactly this skill set.

Where these skills are in demand

AI Solutions Architect

Designing AI systems end to end - models, vector databases and orchestration.

AI Product Manager

Framing AI problems around business metrics and acceptance criteria.

Technical Program Manager

Scoping, planning and delivering AI initiatives within real constraints.

AI Governance Officer

Understanding how AI is built and tested - the foundation for governing it.

AI Engineer

Building RAG pipelines and prototypes that teams can put in front of users.

Machine Learning Engineer

Choosing between predictive ML, RAG and fine-tuning for each use case.

LLM Application Developer

Knowledge assistants and "company brain" tools grounded in internal data.

AI Consultant

Assessing feasibility and making the business case for AI solutions.

The honest comparison

What you're actually paying for

Every topic in this course exists free somewhere on the internet. Nobody is pretending otherwise. What a course adds is the order you learn it in, the practice that makes it stick, and something to show at the end.

What you get
This course
DIY with free content
Business and technicalBoth halves of the decision
Problem framing, KPIs and architecture
Code tutorials with no business context
Decision frameworksWhen to use what
ML vs GenAI, RAG vs fine-tuning, build vs buy
Whatever the latest post recommends
Practice with feedbackLearning that sticks
15 assignments and quizzes
Nobody checks your work
A working prototypeSomething to demo
End-to-end RAG MVP with a Streamlit UI
Notebooks nobody else can use
Proof for recruitersShareable, on LinkedIn
Shareable Coursera certificate
Nothing to show
07 Where to go next

More from the Managing AI Systems specialization.

This course stands on its own - but it's also the first step of a three-course path. Enrolling automatically enrolls you in the specialization.

You're here Course 1

AI Systems Design: RAG Pipelines and LLM Architecture

Problem framing, the modern AI stack, data strategy and a RAG prototype.

≈ 2 weeks · 4 modules
Back to top
Course 2

MLOps and LLMOps: Deploying and Scaling AI in Production

CI/CD for ML and LLM systems with GitHub Actions and MLflow, and optimized inference.

≈ 2 weeks
View course
Course 3

AI Risk and Compliance: Audit and Governance Foundations

Red teaming, the EU AI Act and the NIST AI RMF across the AI lifecycle.

≈ 2 weeks
View course
Related RAG and LLM courses from Board Infinity
Build a RAG app

Build an AI-Powered Document Summarizer & Q&A System

A complete, production-oriented RAG system for summarizing and querying documents.

View course Go deeper on RAG

RAG from Scratch: Build a Knowledge-Powered Chatbot

Document ingestion, chunking and grounding, built into a deployable chatbot.

View course Add agents

RAG Systems & Agentic Workflows with Pinecone and LangGraph

Vector databases, stateful LangGraph agents and observability.

View course

Board Infinity runs 270+ courses and 57 specializations on Coursera - browse the full catalogue.

Know exactly what you're signing up for.

4 modules · 36 lessons · 15 assignments · free PDF

08 FAQ

Common questions

Everything learners ask before enrolling, answered plainly.

About the course

Basic familiarity with AI/ML concepts is helpful but not mandatory. The course starts with foundational frameworks before moving to hands-on implementation.

It's designed for PMs, TPMs, AI governance officers and engineers alike. Module 1 specifically covers career scope and business-aligned AI problem framing.

A simple RAG chain in Module 2, and an end-to-end RAG prototype - a "company brain" - with an interactive Streamlit UI in Module 4, which you then test and refine from its logs.

LLMs and embedding models, ChromaDB and FAISS for vector search, LangChain and LlamaIndex for orchestration, and Streamlit for the prototype UI.

Coursera lists it at about 2 weeks at 10 hours a week, across 4 modules. It's fully self-paced, so you can spread it out around work.

Career & certificate

Roles such as AI Solutions Architect, AI Product Manager, Technical Program Manager and AI Governance Officer.

When you enroll in the course, you get access to all of the courses in the specialization, and you earn a certificate when you complete the work. Your electronic certificate is added to your Accomplishments page, where you can print it or add it to your LinkedIn profile.

To access course materials and assignments and earn a certificate, you purchase the Certificate experience when you enroll. Eligible learners may be able to start with a free trial, and some may see a Full Course, No Certificate option.

Yes. In select learning programs you can apply for financial aid or a scholarship if you can't afford the enrollment fee. If it's available, you'll find a link to apply on the Coursera course page.

Want someone reviewing your AI design with you?

Board Infinity 1:1 coaching pairs you with a practitioner while you take the course.

See 1:1 coaching

Your company's AI brain is four modules away.

Four modules. Fifteen assignments. A RAG prototype you designed and built yourself - and a certificate to show for it.

Enroll on Coursera

AI Systems Design: RAG Pipelines and LLM Architecture · by Board Infinity · offered on Coursera

AI Systems Design Enroll