Agentic Workshop
From binary to agents — how software gets built, and where AI takes it next.
Arrow keys, or swipe →
A quick show of hands.
No wrong answers — this is just where the tour starts from for you. Tap the ones that apply.
The evolution of software.
Every layer of software is built on the one beneath it. Four of them, stacked — from the machine at the bottom to plain language at the top.
Binary
Every computer, at bottom, only understands two states: on and off — written as 0 and 1, bits grouped into bytes. Every program that has ever run was binary at some point.
Low-level languages
Assembly and C let you write closer to what the machine actually does. You manage memory yourself — no safety net. Fast and precise, but every detail is your responsibility.
High-level languages
Languages that read closer to English than to machine code. The language handles memory, so you focus on logic. A trade: some speed and control, for a lot of speed writing it.
Vibe-coding
Describe what you want in plain language; the AI writes the code — and decides the syntax, the structure, even the language. Another layer of abstraction: the same evolution, one step further.
Four layers, one evolution.
Each layer hides the one below it. Vibe-coding doesn't replace the others — it sits on top of them, the way Python sits on top of C, and C sits on top of binary.
“Maybe in ten years, talking to an AI will feel just as outdated. Maybe we'll just think, and a product appears.” — on where this evolution might go
Building an agent.
A program that can reason, decide, and act — not just answer. It can call tools: search the web, run code, send a transaction.
What is an agent?
…then observe again
Choosing your stack
Run it on your machine
Qwen 2.5 (7B) or Hermes 3 (8B) through Ollama, 4-bit quantized (Q4_K_M GGUF) to fit in ~4.3–4.7 GB. Windows, Linux and macOS, with an 8 GB RAM floor. On Apple Silicon (M4), Metal acceleration gives 40–50+ tokens/sec.
Call a model in the cloud
No local compute needed. OpenRouter gives API access to dozens of models, including free tiers — a good fallback when your machine isn't powerful enough.
Running it
$ ollama run qwen2.5:7b-instruct-q4_K_M pulling manifest... done $ ollama run hf.co/NousResearch/Hermes-3-Llama-3.1-8B-GGUF:Q4_K_M >>> ready
$ export OPENROUTER_API_KEY=sk-... $ agent.model = "openrouter/qwen-2.5-7b-instruct" >>> connected
Watching the agent think
Every step is visible: what the agent decided, which tool it called, and what came back.
> user: what's my Sui balance? > agent: calling get_balance() > tool: 12.4 SUI > agent: You have 12.4 SUI.
Agentic wallets on Sui.
The agent reasons in natural language, but every action it takes is a real, signed transaction on Sui.
From request to on-chain action
What we're building
The agent proposes the transaction in plain language; you stay the one who signs.
Thank you — questions? Let's build.
Agentic Workshop — a session at Sui Basecamp 2026 by BIAR. Marina Bay Sands, Singapore · Oct 7–8.
Take it with you.