Your Sunday, upgraded. Free live AI class · 5 PM ISTReserve your seat
Join free class
Build / AI Applications
PRODUCT BUILD · PROPOSED

Build a Knowledge assistant

An internal assistant over company knowledge with permissions and evaluation.

PRODUCT BUILDING COURSE12 hours
Choose how to learn

Live, interactive classes with time for every learner's questions and code reviews.

Request this batch A 5–10 learners batch starts once enough learners ask. You confirm and pay only when dates are set.
AI ApplicationsAdvanced4 modulesAI featureEnrolment not yet open
Build it live on SundaySee the learning steps
AT A GLANCE

What this build involves.

Prerequisites

Python basics and calling a REST API. No machine learning background needed.

Difficulty
Advanced
Build type
Complete product build
Formats
Webapp, Browser extension, API
Build stages
Concept → POC → MVP → Production implementation
Learning formats
Guided series, Mentored build
Stack
Python · FastAPI + LangChain · PostgreSQL + pgvector · Docker

Only need the code? The knowledge assistant source code is sold separately for ₹1,799. Request the source code — availability is confirmed before you pay.

DEMO

See what you will build.

A recorded walkthrough of the finished knowledge assistant is shared when this build opens. Register interest to be told first.

FEATURE CHECKLIST

Included modules.

  • Connectors
  • Retrieval pipeline
  • Access control
  • Evaluation set
  • Authentication and user roles
  • Deployment to Docker
LEARNING STEPS

Seven units, from concept to improvement.

  1. 01

    Concept

    Who uses a knowledge assistant, what problem it solves and the use cases you will support.

  2. 02

    Fundamentals

    Python, FastAPI + LangChain and PostgreSQL + pgvector, plus the data model and architecture for this product.

  3. 03

    POC

    A working demonstration of connectors.

  4. 04

    Feature build

    Retrieval pipeline, Access control, Evaluation set, with authentication and data validation.

  5. 05

    Complete project

    All 4 modules connected into one usable webapp.

  6. 06

    Production implementation

    Automated tests, security review, deployment to Docker, monitoring and documentation.

  7. 07

    Improvement challenge

    Measure answer quality with an evaluation set and cut cost per answer by half.

WAYS TO LEARN THIS BUILD

Start small or go all the way.

01

Free micro lesson

The data model behind connectors

02

Microcourse

Build the connectors module

03

Live workshop

Build retrieval pipeline live, step by step

04

Paid build series

Connectors → Retrieval pipeline → Access control → Evaluation set

05

Mentorship

Review of your architecture and code

BUILD IT LIVE

Build the core of this product with Ajay, live.

The Sunday workshop “Build an AI Search Application” builds the heart of this product step by step. Join it first, then extend it into the full build below.

See dates and register
Workshop
Build an AI Search Application
Date
Sunday, 3 October 2027
Sessions
English 11 AM–2 PM · Hinglish 8 PM–11 PM IST
You leave with
Leave with a working ai search application with sample inputs and verification notes.
REPOSITORY & REVIEW

What you get, and how it is checked.

Repository access

Enrolled learners get access to the starter repository: project structure, database schema, sample data and a setup guide, plus a reference solution for each unit.

Review criteria

  • Every feature on the checklist works end to end
  • Data model and architecture are explained in the README
  • Inputs are validated and access is restricted to the right users
  • Key flows are covered by automated tests
  • The app is deployed and reachable from a public URL
  • Interfaces meet basic accessibility checks

Improvement challenge

Measure answer quality with an evaluation set and cut cost per answer by half.

Project mentorship

Architecture, debugging and implementation reviews with Ajay.

Explore mentorship
AFTER YOU FINISH

A certificate, and support for what comes next.

Finishing this build is planned to include the support below. It opens with the first batch; enrolment pages confirm what is included before you pay.

Completion certificate

Issued with a unique ID once your final project passes review. It records what you built, not just attendance.

Certification policy

Job-readiness support

Portfolio and GitHub review, resume feedback, a mock interview and referrals when suitable openings come up. No placement guarantee.

Career mentorship

Freelancing support

Guidance on pricing, proposals and your first client project, plus a review of your freelance profile and portfolio.

Build client-style projects

Startup community

An invite to the Innovate with Ajay community to meet builders, find co-founders, join hackathons and get feedback on product ideas.

Explore the community
YOUR NEXT LEARNING STEP

Want to learn this?

Share your interest. We’ll confirm availability, batch dates and the fee separately, so you can decide when the details are ready.

  • Your course is already selected
  • Just your name and email to get started
  • No payment or booking commitment
Looking for a different course? Tell us →

Get availability updates.

YOU’RE INTERESTED INBuild a Knowledge assistant
Add learning preferences Optional

No payment now. Availability, dates and the fee are confirmed separately. We’ll use these details to reply. Privacy policy

Prefer a quick chat?Ask on WhatsApp (opens a new tab)
RELATED BUILDS

More in AI Applications.

Tech, Junior

Website AI representative

What you build
An assistant on a website that answers visitor questions from the site's own content and hands off to a human.
Prerequisites
Python basics and calling a REST API. No machine learning background needed.
Stack
Python · FastAPI + LangChain · PostgreSQL + pgvector
  • Content ingestion
  • Grounded answers
  • Lead capture
  • Human handoff
Final deliverableA working website AI representative with content ingestion, grounded answers, lead capture and human handoff, tested and deployed to production.
Intermediate4 modulesWebapp, Browser extension, API
From₹588₹1,18851% off· 12 hours
View projectRequest
Tech, Junior

Document Q&A

What you build
Upload documents and ask questions with answers that cite the source.
Prerequisites
Python basics and calling a REST API. No machine learning background needed.
Stack
Python · FastAPI + LangChain · PostgreSQL + pgvector
  • Document upload
  • Chunking & embeddings
  • Cited answers
  • Conversation history
Final deliverableA working document Q&A with document upload, chunking & embeddings, cited answers and conversation history, deployed as an MVP.
Beginner4 modulesWebapp, Browser extension, API
From₹490₹99051% off· 10 hours
View projectRequest
Tech, Junior

Meeting notes assistant

What you build
Turn meeting recordings into summaries, decisions and action items.
Prerequisites
Python basics and calling a REST API. No machine learning background needed.
Stack
Python · FastAPI + LangChain · PostgreSQL + pgvector
  • Audio upload
  • Transcription
  • Summaries & action items
  • Task export
Final deliverableA working meeting notes assistant with audio upload, transcription, summaries & action items and task export, tested and deployed to production.
Intermediate4 modulesWebapp, Browser extension, API
From₹588₹1,18851% off· 12 hours
View projectRequest
NOT SURE WHERE TO START?

Learn the skills first, then build.

Pick up Python and FastAPI + LangChain in the course catalogue, or follow a full role path.

Browse courses
Chat with Ajay