​Build Your Own CLI Coding Agent from Scratch

This is a 2.5-hour hands-on workshop where we will build a small but practical CLI coding agent from scratch in Python. We'll start with a basic LLM loop and iteratively add capabilities: loading context files, reading files, writing files, and executing shell commands. A core focus of this session will be observability. We'll wire up an event log so we can see exactly what the agent is doing at all times.

All of us use tools like Claude Code or Codex daily, but we likely don't understand what's actually happening inside them. The goal here is to demystify that. We'll see that the core of what makes these tools effective is surprisingly simple: an LLM in a loop with a few tools, not the complexity around it. And unlike training frontier models, building a capable coding agent is something any engineer can do. The simple agent we build in this session won't be just a toy; it can genuinely become your daily driver for work.

The session is mostly live coding with minimal lecture. By the end, you'll have a working coding agent that you built yourself, that you understand completely, and that you can customize however you want.

​Prerequisites

​This is a technical, fast-paced workshop designed for engineers who are comfortable with code. To ensure we can focus on the logic rather than environment setup, please make sure you have the following ready before we begin:
- Python 3.12 and uv installed.
- Comfortable with Python (this is not a beginner workshop).
- An Anthropic API key with ~$10 of credits. We will standardize on Anthropic's API for this workshop.
- Willing to write code manually. We'll disable Copilot and other AI assistance and type every character manually like cavemen.

​Agenda

​18:00 Doors open

​18:30 - 20:30 Workshop

​20:30 - 21:00 Networking

​21:00 Doors close

​Instructor: Prashant Anand

​Bio:

​Organizers

​Prashant Anand: TBD

​​​​​Ilya Kulyatin: Fintech and AI entrepreneur with work and academic experience in the US, Netherlands, Singapore, UK, and Japan, with an MSc in Machine Learning from UCL.

​Supporters

​​​Tokyo AI (​​​TAI) is the biggest AI community in Japan, with 4,000+ members mainly based in Tokyo (engineers, researchers, investors, product managers, and corporate innovation managers).

​Value Create is a management advisory and corporate value design firm offering services such as business consulting, education, corporate communications, and investment support to help companies and individuals unlock their full potential and drive sustainable growth.

​​Privacy Policy

​​We will process your email address for the purposes of event-related communications and ongoing newsletter communications. You may unsubscribe from the newsletter at any time. Further details on how we process personal data are available in our Privacy Policy.