purely.ai

Learn Advanced
AI-Assisted Coding

Live cohort for engineers who can code with AI but cannot yet explain, debug or defend what it produces.
Next Cohort: 6th Dec 2026

“It was a perspective shift in how I work.”
Chinmay Virmani · Founder (Chinsco Studio, Game Dev.)

The AI frontier is jagged, we help you navigate it

About the instructor

Portrait of Dipanshu Rose, the instructor.

Dipanshu Rose

9 years in AI, ML, NLP and automation. Worked with Gartner, NSE, and others, built BerryViral.com, and mentored dozens of engineers and interns into shipping work they can stand behind.

linkedin.com/in/dipanshurose →

What you can do after eight weeks:

Explain every line
with AI off
Spot the bad diff
before you run it
Own your work
maintain with ease and speed
Interrupt or skip
knowing when not to use AI
Orient in an hour
on any new codebase

Learn skills that will help you for decades

what other courses give youother courseswhat the purely.ai cohort gives youpurely.ai
a workflowwhat to do when it breaks
a prompt packbetter ways to frame YOUR problems
a demojudgment that outlasts trending tools
a chat that closes with the courselifetime access to the community

The curriculum

  1. week 1
    Zero to hero: AI-assisted coding
    how we got here · problem formulation · model selection · MCP/tools/skills
  2. week 2
    Build something real with AI: Project 1
    guided build · ship by Friday
  3. week 3
    Making AI trustworthy with tests & automated verification
    tests · AI loops · reading diffs · when to interrupt
  4. week 4
    The art of debugging & handling unfamiliar code
    peer review · unfamiliar code · first principles · git · 1:1 Q&A
  5. week 5
    Advanced prompting and cost management
    prompt iteration · cost · context mgmt · real tools
  6. week 6
    Code you can leave behind: Security, Reviews, Maintenance
    AI-made patterns · security · code review · failure modes · legacy
  7. week 7
    Capstone Project: Software meets the real world
    ambiguous spec · problem formulation · async build
  8. week 8
    Keeping up with AI, without drowning in noise
    AI-off review · jagged frontier · inference · quantization · finetuning · wrap-up
  9. demo day
    Demo day
    present your work · live Q&A · what you'd ship next

How a cohort runs

A timeline of eight blocks on a line — weeks 2 and 7 highlighted, the eighth cut off at the frame's edge.
Cohort Size

Under 10 members, so everyong gets attention, quality reviews, and 1-on-1 guidance from the instructor

Format

Live cohort sessions of 2hrs each, every Sunday afternoon for 8 weeks

Length

8 weeks + a demo day to showcase what you made

Time Commitment

Roughly 6 to 8 hours per week. Live sessions plus async build. Built to run alongside a full-time job.

You need

Your own laptop, your usual editor, and an AI subscription you already have.

It's a proven system, take it from past mentees:

01 / 12

Working with Dipanshu was a perspective shift in how I work. He has never handed me answers directly, always pointed me in directions where I could find these solutions myself.

Chinmay Virmani
Founder (Chinsco Studio, Game Dev.)

We keep track of everything in AI, so you don't have to.

models shipping, moves worth knowing — updated as it happens

Whatever you build, we're here for you.

The skills are about the loop between you and the AI, not a specific platform/framework. Bring your favourite stack.

A row of abstract build targets — browser, phone, controller, terminal, database, neural net, server, chip — threaded together by one yellow line.
webiOS / Androidgame devsystemsdata / MLbackendembeddedanything else

Before you enroll.

Eight weeks from "it works" to "I know how."

Next cohort starts 6th Dec 2026. Enroll now, pay online, and start with your cohort. Limited seats.

₹15,000
one payment / all eight weeks
Enroll Now →