Agentic AI Engineer & Technical Instructor

THE OPPORTUNITY

Build the Agentic AI learning track

Chaduvuko is looking for an Agentic AI Engineer & Technical Instructor who can do more than explain frameworks. We want someone who can design, build, debug, evaluate, and teach modern AI agent systems — and turn that expertise into a structured learning journey that takes learners from strong foundations to production-ready implementations.

This role combines AI engineering, curriculum design, technical teaching, live coding, and educational content creation. You will help define and own Chaduvuko’s Agentic AI & Automation track across our learning platform and YouTube channel.

This is not a traditional classroom trainer role. You should be comfortable answering both “How do we build this?” and “Why should we design it this way?”

WHAT YOU WILL DO

Own the learning experience

You will take ownership of the Agentic AI curriculum from architecture through delivery. Your work will include:

Designing a complete beginner-to-advanced Agentic AI learning roadmap.
Defining prerequisites, modules, lesson objectives, projects, practice exercises, and assessments.
Ensuring foundational concepts are taught before advanced orchestration and production topics.
Creating clear English technical lessons, demonstrations, diagrams, and live-coding walkthroughs.
Building realistic agent systems learners can reproduce independently.
Creating and maintaining source repositories, setup instructions, exercises, assignments, and supporting material.
Teaching not only successful implementations, but also failure cases, debugging, trade-offs, and architectural decisions.
Working with Chaduvuko’s technical reviewers and editors before content is published, and maintaining technical accuracy as tools and frameworks evolve.

The complete curriculum will be reviewed and approved before production begins.

CURRICULUM SCOPE

What you will teach

The track will progress from essential foundations into production-level agent engineering.

Foundations

Python for AI applications
APIs and JSON
LLM fundamentals
Prompt engineering
Structured outputs
Function and tool calling
Basic AI application architecture

Retrieval and knowledge systems

Embeddings and vector databases
Semantic search
Retrieval-Augmented Generation (RAG)
Retrieval quality, chunking, and indexing strategies
Grounding and hallucination reduction
Agentic RAG architectures

Agent engineering

Agent loops and state management
Short-term and long-term memory
Planning, tool selection, and workflow control
Error handling, retries, and conditional routing
Human-in-the-loop approval
Long-running workflows

Agent frameworks and orchestration

LangGraph, LangChain, LlamaIndex, Semantic Kernel, or comparable frameworks
The underlying engineering principles — the course must not depend on a single library

Model Context Protocol (MCP)

MCP architecture: clients, servers, tools, resources, and prompts
Building MCP servers and connecting agents to external systems
Authentication, permissions, and secure MCP usage

Multi-agent systems

Supervisor/worker patterns, delegation, and routing
Specialized agents, shared state, and agent communication
Coordination and failure recovery
When not to use a multi-agent architecture

Production AI engineering

Agent evaluation, evaluation datasets, and regression testing
Tracing, observability, and monitoring
Guardrails, prompt-injection risks, tool permissions, and authentication
Data privacy and security boundaries
Latency, model selection, cost optimization, reliability, and deployment

Real-world projects

An AI research assistant or retrieval-based knowledge assistant
A customer-support or data-analysis agent
Tool-using workflow automation and MCP-enabled assistants
A human-in-the-loop business workflow or multi-agent application
Each with architecture, implementation, testing, debugging, evaluation, security, and deployment where appropriate

WHAT YOU BRING

Skills and experience

You should have strong hands-on experience building modern AI applications and be able to explain your engineering decisions clearly.

Required skills

Strong Python programming and REST/API integration
LLM application development, tool/function calling, and structured outputs
RAG, embeddings, and vector databases
Agent orchestration and application architecture
Debugging, testing, and evaluation
Git and GitHub, and deployment fundamentals

You should also be comfortable reasoning about:

Agent state, memory, and planning
Tool permissions and human review
Reliability, security, and cost
Multi-agent trade-offs

Teaching ability

Technical depth alone is not enough. You should be able to:

Explain difficult concepts in clear English, and teach a beginner without oversimplifying the engineering.
Sequence concepts logically and explain code while writing it.
Use diagrams where architecture matters, and demonstrate failures and debugging.
Connect individual concepts into complete systems.
Record clear screen demonstrations with readable code and clean audio.

Previous teaching experience is valuable, but not mandatory if you can demonstrate strong communication ability.

Preferred experience

We would especially value candidates with experience in:

LangGraph or similar orchestration frameworks, and MCP integrations
Production RAG applications, multi-agent systems, and LLMOps
AI evaluation, tracing, and observability
Cloud deployment, Docker, FastAPI, and CI/CD
AI application security
Technical writing, workshops, mentoring, or YouTube / online-course creation

Public repositories, technical write-ups, and working applications are strong signals.

How we think about experience

We generally expect 3–7+ years of relevant engineering experience, but we do not require a specific number of years working in “Agentic AI.” The field is evolving too quickly for that to be a useful filter.

We care more about what you have built, how well you understand it, how you reason about architecture, and how clearly you can teach it. Strong candidates with less traditional experience but exceptional Agentic AI work are encouraged to apply.

HOW WE BUILD CONTENT

Lesson standards

Each lesson should have:

A clear learning objective and defined prerequisites
A structured lesson outline and supporting resources
Runnable code where applicable, with setup instructions
Practice or assessment where relevant

Typical target durations:

Concept lessons: 8–15 minutes
Technical demonstrations: 15–25 minutes
Project lessons: 15–30 minutes

Long topics should be divided into logical lessons rather than compressed into oversized videos. Lesson count follows the curriculum and learning dependencies, not an arbitrary video quota.

OUTCOMES

What success looks like

Learners can follow the track without prerequisite gaps.
Code examples work and can be reproduced.
Projects reflect realistic engineering scenarios, and architecture decisions are explained clearly.
Lessons remain technically accurate, and content is delivered according to agreed milestones.
Review feedback is incorporated professionally.
Learners progressively move from understanding concepts to building complete systems.

YouTube views are useful, but they are not the sole measure of instructor performance.

THE PROCESS

How we hire

01
Application reviewWe review your professional background, projects, portfolio, GitHub, and any relevant technical content.
02
Technical conversationWe discuss one or more systems you have built — architecture, trade-offs, failures, evaluation, security, deployment, and your personal contribution.
03
Teaching demonstrationYou explain a relevant concept in a short technical demonstration. An existing video is fine; a studio-quality recording is not required.
04
Paid pilotShortlisted candidates may complete a small pilot lesson or module. Scope, payment, deliverables, and review criteria are agreed in writing before any substantial trial work begins.
05
Final reviewWe assess technical accuracy, teaching clarity, engineering depth, code quality, communication, and responsiveness to feedback.
06
Agreement and onboardingBefore the main engagement begins we confirm engagement type, compensation, scope, milestones, deliverables, revision expectations, confidentiality, content ownership, and termination terms.

PAY & ENGAGEMENT

Compensation

For suitable full-time engagements, Chaduvuko currently expects a compensation range of approximately ₹60,000–₹1,20,000 per month. Final compensation depends on technical depth, teaching ability, experience, responsibilities, and scope.

We may also engage instructors through course-based, module-based, part-time, or milestone-based agreements. Contract compensation is agreed against the approved scope before work begins.

There are no recruitment fees, required purchases, or paid subscriptions to apply.

READY TO APPLY?

What to include

Use the subject “Application — Agentic AI Engineer & Technical Instructor” and include:

Your resume or professional profile.
Relevant Agentic AI / RAG / LLM project links, and GitHub or portfolio where available.
A brief description of your contribution to at least one project.
Your preferred engagement type, availability, and expected compensation.
A teaching sample, technical video, workshop recording, or technical article if you have one — this is not required to apply.

Candidate privacy

Please submit only information relevant to the position. Do not send identity documents, bank information, or other unnecessary sensitive records during the initial application stage. Application information is used for recruitment and candidate evaluation for this role. Questions go to hello@chaduvuko.com.

Questions about the role or application? hello@chaduvuko.com