Course Details

Courses Descriptions

MODULE OBJECTIVE

What you'll achieve

Package AI apps into Streamlit UIs, deploy on public cloud (AWS/GCP), implement LLMOps monitoring with LangSmith, establish CI/CD for AI, and understand LLMOps as a discipline.

SECTION 01 — Streamlit for AI Apps

Rapid UI for LLM and ML applications

• Streamlit fundamentals: layout, widgets, session state
• Building chat interfaces for LLM agents
• File upload, rich output, multi-turn state
• Connecting Streamlit to LangChain/LangGraph

SECTION 02 — Cloud Deployment

Taking apps live on public cloud

• Docker containerisation for AI apps
• Streamlit Community Cloud: free public hosting
• AWS: EC2, ECS, Lambda + API Gateway
• GCP: Cloud Run, Vertex AI deployment

SECTION 03 — LLMOps Fundamentals

Operating LLMs in production

• What is LLMOps? Comparison with MLOps
• Prompt versioning and experiment tracking
• LangSmith: tracing, evaluation, datasets
• Cost monitoring: token usage and API budgets

SECTION 04 — Observability & Evaluation

Measuring quality over time

• Logging agent traces and intermediate steps
• LangSmith evaluation runs and quality scoring
• Regression testing for quality drift detection
• User feedback loops and RLHF signal collection

SECTION 05 — CI/CD for AI Systems

Shipping AI safely and continuously

• Git-based workflow for prompt and code changes
• Automated test suites for LLM outputs
• Blue/green deployment for zero-downtime updates
• Program summary: the complete AI mastery stack

KEY TAKEAWAYS

1. Streamlit turns a notebook into a web app in minutes

2. Docker + Cloud Run is the fastest path to a production URL

3. LLMOps = MLOps + prompt versioning + trace observability

4. LangSmith tracing is non-negotiable for debugging agents

5. CI/CD for AI means testing prompts, not just code

6. Full stack complete: concept → code → deploy → monitor