All coursesEngineering

Agent-First Software Engineering

4.7 (124)

Vignesh Mohankumar

Staff Eng, currently AI Consultant

Ship production code 10× faster with the workflows staff engineers actually use

You’ll see us build real software using the exact workflows we use in our own consulting work. This is not a vibe coding course.

We built this accelerator because we’re engineers first. For the past few years, we’ve been using AI tools to ship production features that used to take weeks in days. We started by sharing these workflows privately with the senior engineers we worked with. We created this accelerator to scale the knowledge to hundreds of developers.

You will learn both:

  • Synchronous workflows — coding live with AI, structuring state and context, getting clean and correct code out of models.

  • Asynchronous workflows — background jobs, automated agents, Slack triggers, multi-step processes that run while you focus on something else.

And we cover everything that surrounds coding, too:

  • Design-doc generation from meeting notes

  • Debugging with AI

  • PR review workflows

  • Custom commands, sub-agents, rules

By the end of the accelerator, you’ll have a repeatable engineering framework you can apply to any codebase, stack, and dev environment. And you’ll walk away having shipped a real feature using these methods.

Trusted by 200+ developers.

What you’ll learn

Move from ad-hoc AI usage to a clear, repeatable workflow for shipping production software with confidence.

  • Learn the Explore → Plan → Build framework used by senior engineers to turn vague ideas into reviewable, low-risk plans.

  • Practice incremental code generation and stricter review habits so AI output is production-safe.

  • Learn how to interrogate systems, architecture, and dependencies using AI without hallucination-driven thrash.

  • Replace days of trial-and-error with structured exploration and informed decision-making.

  • Learn when async delegation helps — and when it creates more problems than it solves.

  • Run background research, refactors, and audits while maintaining full human ownership.

  • Adopt workflows that produce smaller, clearer diffs that are easier to review and reason about.

  • Learn how to review AI-generated code more skeptically than human-written code.

  • Establish a common language for planning, ownership, and review that managers and engineers can align on.

  • Prevent AI adoption from devolving into tool-driven chaos or hype cycles.

Learn directly from Vignesh

Vignesh Mohankumar

Vignesh Mohankumar

Contact

Founding Engineer @ Multiple Exited Startups

Who this course is for

  • Engineers who want consistent, production-safe workflows, not guesswork or oversized diffs.

  • Team leads and managers responsible for AI adoption who need practical standards, not hype or tool comparisons.

  • Engineers tired of vibe coding and social-media advice who want real workflows from someone shipping code daily.

Prerequisites

  • Experience shipping production software

    You should be comfortable building and maintaining features in your primary language or stack so you can focus on workflow, not syntax.

What's included

Vignesh Mohankumar

Live sessions

Learn directly from Vignesh Mohankumar in a real-time, interactive format.

Lifetime access

Go back to course content and recordings whenever you need to.

Community of peers

Stay accountable and share insights with like-minded professionals.

Certificate of completion

Share your new skills with your employer or on LinkedIn.

Maven Guarantee

This course is backed by the Maven Guarantee. Students are eligible for a full refund up until the halfway point of the course.

Course syllabus

Week 1

Mar 25—Mar 27

    Day 1 — Fundamentals of AI Software Engineering

    • Mar

      25

      Fundamentals

      Wed 3/254:00 PM—6:00 PM (UTC)

    • Mar

      25

      Live Coding Example

      Wed 3/256:30 PM—8:30 PM (UTC)

    • Mar

      25

      Wrap-up / Discussion

      Wed 3/258:30 PM—9:00 PM (UTC)

    Day 2 — Advanced Workflows

    • Mar

      26

      Coding Agent Tooling

      Thu 3/264:00 PM—6:00 PM (UTC)

    • Mar

      26

      Coding Agent Infrastructure

      Thu 3/266:30 PM—8:30 PM (UTC)

    • Mar

      26

      Wrap-up / Discussion

      Thu 3/268:30 PM—9:00 PM (UTC)

    Day 3 — Hackathon & Implementation

    • Mar

      27

      Optional: Hackathon/Office Hours Session 1

      Fri 3/2712:00 PM—3:00 PM (UTC)
      Optional

    • Mar

      27

      Optional: Hackathon/Office Hours Session 2

      Fri 3/274:00 PM—7:00 PM (UTC)
      Optional

Free resource

AI Coding Accelerator: How OpenAI uses Codex cover image

AI Coding Accelerator: How OpenAI uses Codex

Codex Evolution & Core Features

Discover how Codex transformed OpenAI's workflow and master the essential features driving productivity

Advanced Prompt Patterns & Techniques

Apply proven prompt strategies and hidden capabilities used by OpenAI engineers in real projects

Integration Best Practices

Implement Codex effectively in your development stack using OpenAI's tested approaches

Schedule

Live sessions

14-20 hrs

Live Coding Session and Hackathons

    • Wed, Mar 25

      4:00 PM—6:00 PM (UTC)

    • Wed, Mar 25

      6:30 PM—8:30 PM (UTC)

    • Wed, Mar 25

      8:30 PM—9:00 PM (UTC)

Frequently asked questions

$1,350

USD

·
Mar 25—Mar 27
Enroll