On-site or online

Agentic Engineering training

A one-day hands-on training for engineering teams who want to build production software with Claude Code.

What this is

Most teams have AI tools installed and nothing has changed. Copilot is autocompleting, ChatGPT tabs are open, but the code still ships at the same pace it did before — sometimes slower. The problem isn’t the model. It’s the workflow around it.

This training is the workflow. One day with your engineers, on-site or online, building real features on your own stack. We wire up the agent, the conventions, and the review gates, and leave you with a process your team keeps using after I leave.

The agent does the typing. You still own the architecture.

What we cover

Claude Code
Context Building
CLAUDE.md
Plan Mode
MCPs
Skills
Subagents
Spec-Driven Workflow
Refined SDLC
CI/CD Pipelines
Code Review

The problem

Your team has AI tools. But nothing has changed.

Copilot is installed. ChatGPT tabs are open. But your team is not shipping faster — and they know it.

Problem 01
Everyone prompts differently
No shared patterns, no consistent output. Each engineer gets different results from the same tools, and nobody trusts what comes out.
Problem 02
AI made things slower
Review cycles doubled. Debugging AI-generated code takes longer than writing it from scratch. The speed promise didn’t land.
Problem 03
Impressive demos, broken production
The AI demo looked great. Then it hit real code, real edge cases, and real users. What works in a playground breaks in production.

The shift

Here’s what changes.

Before

Each engineer prompts their own way
AI experiments that never reach production
No review process for AI-generated code
Different results from the same tools
More time debugging AI output than writing code

After

One shared workflow the whole team follows
AI agents that ship real features, not prototypes
Review gates that catch problems before merge
Consistent output across the entire team
Engineers who choose AI because it’s faster

How it works

From first call to working system.

Phase 1
01
Discovery
I learn your stack, understand how your team works, and we decide what to demo during the training.
30-minute call
Phase 2
02
Training
Online or on-site with your team. I walk through real, working examples on your stack with AI agents.
Full day on-site · 5h online
Phase 3
03
SDLC Playbook
You get a documented dev process your team can follow independently, adapted to how you already work.
Custom for your team
Phase 4
04
Follow-Up
Post-training check-ins to ensure adoption sticks. Direct access for real questions.
2 weeks included

Is this for you?

For builders, not browsers.

If you’re looking for an inspirational AI keynote, this isn’t it. This is a working session where we write code and build systems together.

You run a software company and your team writes code every day
You want to ship faster, not just talk about AI
Your team tried Copilot or ChatGPT but nothing stuck
You learn better by building than by watching presentations
You want a process your team follows after the training ends

Capabilities

Skills your team keeps.

Managing Context Windows

AI agents lose track in large codebases. The Plan / Execute / Clear loop keeps them focused and useful.

Steering AI Agents Reliably

AGENTS.md files, custom skills, and progressive disclosure give you control over what the agent does and doesn’t do.

Planning with PRDs

Break features into chunks that fit a context window. Validate the architecture with a tracer bullet before writing the rest.

Integrating AI into CI/CD

Your pipeline can run AI-powered tests, reviews, and checks automatically. We set that up during the training.

Running Autonomous Loops

Let agents code on their own while you review at checkpoints. Useful for large refactors, test generation, and boilerplate.

Preparing the Codebase

Most repos aren’t set up for AI agents. Small structural changes make a big difference in what the agent can do.

What founders say

“Emran sat with our dev team, understood how we work, and tweaked our entire process around AI tooling. He has a rare ability to cut through the noise and deliver something that actually works.”

Parvez Akther

Founder & CEO, ThriveDesk

“Emran is one of the frontiers in AI-based development in our local tech scene. Working with him was a great experience. Would highly recommend.”

M. Mahbubur Rahman

Co-Founder & CTO, iViveLabs

“What sets Emran apart is his deep understanding of real-world software development. He doesn’t just talk about AI, he shows you how to ship with it.”

Eng. Ahmad Naser

Founder & CEO, Crebsol

Investment

Two ways to run it.

Same training. Same outcome. Pick the format that fits your team.

Online · Live

Remote Training

$699

up to 5 hours, anywhere in the world

Same training, run live over video. Recorded for the team to revisit.

Up to 5 hours, live with your team
1–2 working examples on your stack
AI-powered SDLC playbook, customized
Session recording + shared doc
2 weeks of follow-up over chat

Let’s talk

On-site · In-person

On-site Training

$1,499

+ travel & accommodation, billed at cost

I come to your office. Whiteboards, pair sessions, side conversations — the works.

Full day, on-site at your office
Whiteboard + breakout time with leads
1–2 working examples on your stack
AI-powered SDLC playbook, customized
Pre-training discovery & stack review
2 weeks of follow-up over chat

Let’s talk

Multi-day engagements, recurring sessions, and custom team rates are available — let’s talk.

Who teaches it

Mohammad Emran Hasan

Co-Founder & CEO of Klasio / FigLab

25+ years building software products that have survived real users, real constraints, and real growth. I don’t chase trends; I build systems that stay useful. AI-augmented engineering is no different.

Newsletter

Practical Agentic Engineering.

Real-world lessons from adopting AI agents in software teams. From my own workflow. Every other week. No fluff.