Anyone can call an AI API. You will build systems people rely on — retrieval that cites its sources, agents wired to real business tools, and production LLM backends. Level 1 or equivalent required.
There is a canyon between “I made a chatbot” and “I built the AI assistant a company runs its support desk on.” On one side: demos. On the other: retrieval over real data, agents that operate real tools, answers with citations, and costs you can defend to a CFO. This level takes you across that canyon — the same ground Panabotics covers in client work every week.
PKR 100,000Pakistan$115 / £90International
The PractitionerLevel 2
A practitioner is the person a team turns to when they say “can we get AI to do this — properly?” You will design retrieval pipelines over messy real-world data, wire agents to calendars, databases and inboxes, control hallucinations instead of hoping they don't happen, and hand over work that keeps running after you leave the room. This is the level where AI stops being your hobby and becomes your profession.
Ten weeks from LLM APIs to client-grade systems — production prompting, AI backends, data layers, full-stack chat UX, retrieval that cites sources, tool-using agents wired to real business apps, then deploy and operate what you ship.
Ship one client-grade system end to end — a document-intelligence product, a support agent grounded in company knowledge, an internal knowledge assistant, or an AI feature inside a SaaS workflow — deployed, observable, cost-aware, and defended for certification.
After this level, these are things you quote for — not things you Google:
Panabotics Certified AI Foundations Developer (PCAFD) — or comfortable Python plus at least one shipped LLM API project.
One payment · 10 weeks · 55+ video lessons · 8 hands-on projects
Every PIAI seat is reviewed. Submit once — we guide you from application to selection.
Share your details, LinkedIn, and CV for the program you want to join.
Our PIAI team reviews fit, readiness, and prerequisites for your chosen track.
A short conversation to confirm goals, commitment, and whether this is the right level.
Accepted applicants receive an offer with next steps to enroll and start.
Fees, schedule, language, prerequisites, and what the credential is worth — answered plainly, including the answers that might talk you out of applying.
You need Level 1 or equivalent experience. If you already write Python comfortably and have shipped at least one project using an LLM API, you can apply directly — the admissions interview exists to confirm that. If you have never called an AI API from code, start with Foundations instead; this level moves too fast otherwise.
No, and that distinction matters. This is applied generative AI engineering: you use existing frontier models rather than training your own. There is no regression, classification, or neural-network training here. What you learn instead is how to build production systems around LLMs — retrieval, agents, evaluation, and cost control.
RAG — retrieval-augmented generation — is how you make an AI answer from a company's own documents instead of only its training data. Module 6 is dedicated to it, and you will build retrieval pipelines with embeddings and vector databases that return answers with citations, then measure whether that retrieval is actually any good.
The fee is PKR 100,000 in Pakistan, or $115 / £90 internationally, covering all 10 weeks, 55+ video lessons, 8 hands-on projects, and the PCAD certificate assessment. This level is payable in full — installment plans are not offered at Level 2 — and fees are non-refundable once enrollment is confirmed.
Ten weeks, with live classes every Saturday and Sunday from 3:00pm to 6:00pm Pakistan Standard Time — 6 live hours per week and 60 hours in total. All classes are online, so you can join from Lahore, Karachi, Islamabad, or anywhere with a reliable connection. Expect substantial work between sessions.
You work across multiple providers rather than being locked to one: OpenAI, Anthropic, Gemini, Groq, and OpenRouter, including provider routing and fallback. On the engineering side you build with Python, FastAPI, Pydantic, vector databases, and full-stack chat interfaces, then deploy and operate what you ship.
Eight projects that read as client work rather than tutorials — a multi-model gateway that routes prompts by cost and quality, a versioned prompt system with guardrails, an authenticated AI API with webhooks, a RAG system that cites its sources, and agents wired to real tools like calendars, email, and databases.
It gives you the work that makes those conversations possible — deployed systems, measured retrieval quality, and defensible cost engineering are what buyers and hiring managers actually probe. We do not promise placement, and any provider who guarantees a job for a fee is worth being suspicious of.
Classes are taught in English with Urdu explanation. At this level the vocabulary matters — embeddings, chunking, idempotency, tenancy — so terms stay in English exactly as you will meet them in documentation and code review, while the reasoning behind them gets explained in the language you think in.
Level 2 makes you the person who builds AI systems a business depends on. Level 3 — Agentic AI Engineer — makes you the person who architects and operates them at product scale: multi-tenant system design, production AWS, event-driven backends, MCP integrations, and the evals and security to leave agents running unattended.
You earn the Panabotics Certified AI Developer (PCAD) credential, awarded on assessment rather than attendance. You must ship the capstone and defend it. Its value comes from the fact that it cannot be obtained by watching videos — the systems behind it are the actual portfolio.
Tell us who you are and why you want Panabotics Certified AI Developer. We review every application personally before inviting you to the next step.
10 weeks. 8 hands-on projects. One credential — PCAD. Talk to us and we'll help you decide if this is your level.