AI for Data Teams · Intermediate

AI for Data Teams: LLMs, RAG and Reviewing AI Output

Use AI productively in data work, and know when not to trust it.

A practical course for data professionals on large language models: prompt design for data tasks, retrieval-augmented generation, protecting sensitive data and reviewing AI output rigorously.

Duration
16 hours
Mode
Live online
Next batch
To be announced
Schedule
Sat–Sun, 10am–2pm IST

Who it is for

Data engineers, modelers, analysts and team leads introducing AI into data work.

Prerequisites

Experience in any data role. Basic Python is helpful but not required.

What you will be able to do

  • Explain how LLMs work and where they fail
  • Design prompts for SQL, documentation and data quality tasks
  • Build a simple retrieval (RAG) pipeline over your metadata
  • Set up review gates so AI output is checked before use
  • Keep personal and sensitive data out of AI tools

Course contents · 4 modules · 16 hours

01
LLM basics for data people

How models work, strengths, failure modes, hallucination

3 h
02
Prompting for data work

SQL generation, documentation, data quality rules, STTM drafts

4 h
03
Retrieval and vector search

Embeddings, chunking, RAG over metadata and documentation

4 h
04
Safe use and review

Data privacy, human gates, review checklists, measuring quality; capstone reviewed by the trainer

5 h

Hands-on approach

30% concepts, 70% hands-on labs on realistic business scenarios. AI practice support between sessions; every project is reviewed by the trainer.

Keep exploring
SnowflakeIntermediate

Snowflake Data Engineering Bootcamp

For data engineers and SQL developers who want to build reliable, cost-aware pipelines on Snowflake.

Duration
24 hrs
Mode
Live online
Starts
To be announced
Syllabus · 5 modules · Sat–Sun, 10am–1pm IST
  1. Snowflake architecture and setup · 4h
    Storage, compute and services layers; warehouses; databases and schemas; roles and cost basics
  2. Loading data · 5h
    Stages, file formats, COPY INTO, Snowpipe, semi-structured data with VARIANT
  3. Transformations and ELT · 6h
    Streams, tasks, dynamic tables, MERGE patterns, incremental loads, time travel
  4. Security and governance · 4h
    RBAC design, masking and row access policies, tagging, data sharing
  5. Performance, cost and capstone · 5h
    Clustering, query profile, resource monitors; end-to-end capstone reviewed by the trainer
You will be able to: Design Snowflake databases, schemas and warehouses for a real workload · Load batch and continuous data with COPY, stages and Snowpipe · Build incremental ELT with streams, tasks and dynamic tables · Apply role-based access, masking and cost controls · Review AI-assisted SQL critically before it runs
Fees on requestEnquire for fees and the next batch
View course detailsEnquire
DatabricksIntermediate

Databricks Lakehouse Engineering with PySpark

Build medallion-architecture pipelines on Databricks with PySpark and Delta Lake.

Duration
30 hrs
Mode
Live online
Starts
To be announced
Syllabus · 5 modules · Weekday evenings, 7–9pm IST
  1. Lakehouse fundamentals · 4h
    Workspace, clusters, notebooks, Delta Lake basics, medallion architecture
  2. PySpark for data engineering · 8h
    DataFrames, joins, window functions, UDF trade-offs, performance basics
  3. Delta Lake in depth · 6h
    MERGE, change data feed, schema evolution, OPTIMIZE and Z-ORDER, time travel
  4. Governance and quality · 5h
    Unity Catalog, expectations and data quality checks, lineage
  5. Orchestration and capstone · 7h
    Workflows, parameters, monitoring; capstone pipeline reviewed by the trainer
You will be able to: Build medallion (Bronze, Silver, Gold) pipelines with PySpark and Delta Lake · Handle incremental and historical loads, schema evolution and data quality checks · Govern data with Unity Catalog · Orchestrate and monitor jobs with Databricks Workflows
Fees on requestEnquire for fees and the next batch
View course detailsEnquire
Microsoft FabricBeginner

Microsoft Fabric for Data Engineers

Get productive with Microsoft Fabric: Lakehouse, Warehouse, pipelines and Power BI.

Duration
20 hrs
Mode
Live online
Starts
To be announced
Syllabus · 5 modules · Sat–Sun, 2–5pm IST
  1. Fabric and OneLake · 3h
    Workspaces, items, capacities, OneLake shortcuts
  2. Ingestion · 5h
    Data Factory pipelines, Dataflows Gen2, notebooks
  3. Lakehouse and Warehouse · 6h
    Delta tables, SQL endpoint, Warehouse modelling, T-SQL transformations
  4. Semantic models and reporting · 4h
    Direct Lake, measures, Power BI reports
  5. Copilot and review · 2h
    Where Copilot helps, how to verify its suggestions
You will be able to: Navigate Fabric workspaces, OneLake and capacities · Ingest data with pipelines and dataflows into a Lakehouse · Model a Warehouse and build a semantic model for Power BI · Use Copilot features and check their output before relying on it
Fees on requestEnquire for fees and the next batch
View course detailsEnquire