How much does AI engineering tutoring cost?

Track 01 (LLMs & Prompting) and Track 04 (Automation & Deployment) start at $100/hr SGD. Track 02 (Agent Engineering) and Track 03 (Multi-Agent & MCP) start at $120/hr SGD. All sessions are private 1-to-1 — no group classes.

Do I need a machine learning background to start?

No. You need Python experience and ideally some API familiarity. We teach the LLM and agent concepts from the ground up. If you're coming from a software engineering background, you can be building your first working agent in the first session.

What frameworks do you teach?

LangChain, LangGraph, CrewAI, AutoGen, and the OpenAI Agents SDK are the core frameworks. We also cover MCP (Model Context Protocol), n8n and Make for automation, and deployment stacks including FastAPI, Docker, Modal, and Railway.

Are sessions online or in-person?

Sessions are primarily online, with screen sharing and live coding. We serve students globally — Singapore, Malaysia, Australia, the UK, the US, and all timezones. In-person sessions in Singapore are available on request.

What does a typical AI agent tutoring session look like?

Sessions start with a 5-minute check-in on what you built since last time, then move into teaching new concepts with live code examples, hands-on building in your actual project or a shared environment, and end with a clear task for you to complete before the next session. No slides — everything is live code.

How much does AI engineering tutoring cost?

Track 01 (LLMs & Prompting) and Track 04 (Automation & Deployment) start at $100/hr SGD. Track 02 (Agent Engineering) and Track 03 (Multi-Agent & MCP) start at $120/hr SGD. All sessions are private 1-to-1 — no group classes.

Do I need a machine learning background to start?

No. You need Python experience and ideally some API familiarity. We teach the LLM and agent concepts from the ground up. If you're coming from a software engineering background, you can be building your first working agent in the first session.

What frameworks do you teach?

LangChain, LangGraph, CrewAI, AutoGen, and the OpenAI Agents SDK are the core frameworks. We also cover MCP (Model Context Protocol), n8n and Make for automation, and deployment stacks including FastAPI, Docker, Modal, and Railway.

Are sessions online or in-person?

Sessions are primarily online, with screen sharing and live coding. We serve students globally — Singapore, Malaysia, Australia, the UK, the US, and all timezones. In-person sessions in Singapore are available on request.

What does a typical AI agent tutoring session look like?

Sessions start with a 5-minute check-in on what you built since last time, then move into teaching new concepts with live code examples, hands-on building in your actual project or a shared environment, and end with a clear task for you to complete before the next session. No slides — everything is live code.

AI ENGINEERING & AGENTIC SYSTEMS · SINGAPORE & GLOBAL

Build. Deploy. Automate.
AI agents
that actually work in production.

Private 1-to-1 AI engineering mentorship. Learn to design, build, and ship autonomous AI agents using the tools real teams use in 2026 — LangChain, LangGraph, MCP, CrewAI, and more. Online, worldwide. No group classes. Ever.

Start a conversation →See Pricing

agent.py

from langchain import Agent
agent = Agent(
llm="claude-sonnet",
tools=["web_search", "gmail", "notion"],
memory="vector_store",
mcp_servers=["everything"]
)
result = agent.run(
"Research competitors and draft a summary in Notion"
)
print(result)
 # → task complete. no hand-holding required.

AI Agents · MCP · LangChain · Zero to Production · Principal-led · Remote-first · Singapore · AI Agents · MCP · LangChain · Zero to Production · Principal-led · Remote-first · Singapore ·

Accepting students globally

🇸🇬 SG🇲🇾 MY🇦🇺 AU🇬🇧 UK🇺🇸 US🌐 WORLDWIDE

Students & Clients

Principal-Led

Learning Tracks

Tools & Frameworks

WHAT WE COVER

LLMs & Prompting

10 topics

Prompt engineering · Structured outputs · System prompts · Chain-of-thought · Few-shot learning · Context window management · Temperature & sampling · OpenAI API · Anthropic API · Gemini API

Agent Engineering

12 topics

The agent loop · Tool / function calling · ReAct pattern · Plan-and-Execute · LangChain · LangGraph · Memory architectures · RAG systems · Vector databases · Human-in-the-loop · Debugging agents · LangSmith observability

Multi-Agent & MCP

8 topics

Model Context Protocol (MCP) · Building MCP servers · CrewAI · AutoGen · Agent-to-Agent (A2A) · Orchestrator patterns · Specialist agent design · Agent interoperability

Automation & Deployment

9 topics

n8n workflows · Langflow · Make (Integromat) · Docker · AWS / GCP / Vercel deployment · Environment management · Cost optimisation · Ollama (local models) · CI/CD for agents

Agent Security

6 topics

Prompt injection attacks · Data leakage via MCP · Over-permissioned tools · Agent sandboxing · Output validation · Security audits for AI systems

LEARNING TRACKS

Pick your goal

AI Foundations

Beginner

$100/hr

How LLMs work

Tokens, context, temperature

1–2 sessions

Calling the OpenAI / Anthropic APIs

From Python. Your first API call.

1 session

Prompt engineering that isn't fluff

Structured outputs, chain-of-thought

1–2 sessions

Your first RAG system

Vector database, embeddings, retrieval

1–2 sessions

FULL STACK COVERAGE

LLMs

ClaudeGPT-4oGeminiLlama 3MistralOllama

Frameworks

LangChainLangGraphCrewAIAutoGenPydanticAILlamaIndex

Protocol

MCPAgent-to-Agent (A2A)OpenAI Agents SDK

Automation

n8nLangflowMakeZapier AI

Vector DBs

PineconepgvectorChromaWeaviateQdrant

Observability

LangSmithLangfuseHelicone

Deployment

DockerAWSGCPVercelGitHub Actions

WHO THIS IS FOR

Developers

Adding agentic capabilities to existing products or building greenfield agent systems.

Startup founders

Shipping an MVP with AI agents fast — without a full ML team.

Data scientists

Extending from analytics and ML into production agent pipelines.

Automation engineers

Moving beyond n8n/Zapier into custom, code-first agent workflows.

Product managers

Understanding what agents can and cannot do to spec better products.

Career switchers

Breaking into AI engineering from software, data, or non-technical backgrounds.

NUS · NTU · SMU · SUTD · SIT · SIM · SP · NP · TP · RP · NYP
AI modules · Machine learning assignments · FYP · Capstone projects

ZERO TO PRODUCTION

What it actually looks like

Understand the landscape

LLM fundamentals, tokenisation, model selection, cost vs capability trade-offs.

Prompt engineering

System prompts, few-shot, chain-of-thought, structured outputs, function calling.

First agent

Tool use, ReAct loop, state, basic memory. Something that actually does a task.

RAG pipeline

Embeddings, vector store, retrieval, context injection. Grounded in real data.

Multi-agent systems

Supervisor patterns, handoffs, shared memory, MCP integration.

Observability & eval

LangSmith / Langfuse traces, regression testing, output quality gates.

Production deployment

FastAPI wrapper, Docker, cloud deployment, scheduling, monitoring.

UNIVERSITIES & INSTITUTIONS

We work with students from

UNIVERSITY

NUS · NTU · SMU · SUTD · SIT · SIM

AI modules · Machine learning assignments · FYP · Capstone projects · LLM research · Computer vision · Data science

POLYTECHNIC

SP · NP · TP · RP · NYP

AI fundamentals · Machine learning modules · Practical AI projects · Diploma capstone · Smart systems assignments

ITE

ITE College East · West · Central

AI basics · Python programming · Data analysis · Automation intro · Computing coursework

INTERNATIONAL & PROFESSIONAL

Worldwide · Remote · All levels

Career switchers · Working professionals · Corporate upskilling · Self-study acceleration · Career pivots

PROVEN RESULTS

What students say

From zero to deployed agent in 6 weeks

“I came in knowing Python but nothing about LLMs. Six weeks later I had a production LangGraph agent running in AWS Lambda, with LangSmith traces and everything. The depth of knowledge here is unreal.”

— Wei Jian, Backend Developer · Singapore

Finally understood MCP properly

“I'd read the docs three times and still couldn't get MCP to click. One session here and it made complete sense. We built a working MCP server from scratch in that session.”

— Arjun, AI Engineer · Remote (India)

Our startup shipped faster because of this

“We were burning time trying to make CrewAI work for our use case. Two sessions in and we realised LangGraph was the right choice. Saved us weeks of the wrong architecture.”

— Priya, CTO · Early-stage startup, Singapore

WHY DOKKAEBI LABS

Principal-led, always

Every session is with a senior AI engineer who has shipped production agents. No juniors, no subcontractors.

Tooling-agnostic

We teach the right tool for your problem — LangGraph for complex state, CrewAI for role-based crews, raw API for simplicity.

Production focus

We don't stop at 'it runs in a notebook.' Sessions cover deployment, observability, error recovery, and cost management.

Curriculum is yours

No fixed syllabus. Your stack, your use case, your pace. Sessions adapt session to session.

FAQ

Frequently Asked Questions

Have a question? We've got answers. If you don't find what you're looking for, feel free to contact us.

AI agent tutoring is private 1-to-1 mentorship focused on building, deploying, and operating autonomous AI agents. It's for developers, engineers, data scientists, startup founders, and career switchers who want hands-on expertise in LangChain, LangGraph, MCP, CrewAI, and production agent deployment — not just theory.

Still have questions?

Our team is here to help. Get in touch and we'll respond as soon as possible.

Ready to build?

Tell us your stack and your goal. First session is scoped to you.

Message on WhatsApp Send an enquiry →

Build. Deploy. Automate.AI agentsthat actually work in production.

LLMs & Prompting

Agent Engineering

Multi-Agent & MCP

Automation & Deployment

Agent Security

Pick your goal

AI Foundations

How LLMs work

Calling the OpenAI / Anthropic APIs

Prompt engineering that isn't fluff

Your first RAG system

Developers

Startup founders

Data scientists

Automation engineers

Product managers

Career switchers

What it actually looks like

Understand the landscape

Prompt engineering

First agent

RAG pipeline

Multi-agent systems

Observability & eval

Production deployment

We work with students from

What students say

Principal-led, always

Tooling-agnostic

Production focus

Curriculum is yours

Frequently Asked Questions

Still have questions?

Ready to build?

Build. Deploy. Automate.
AI agents
that actually work in production.