Used in our own operations
Research and bilingual publishing
We research primary sources, explore their business implications, and use AI in Japanese and English article production and publication checks.
Research that informs a business decision.
Home RESEARCH & PRACTICE
RESEARCH & PRACTICE
Test it in our business. Share what we learn.
Our own workflows, articles based on international primary sources, and public development resources. We distinguish recorded experience from ideas still being tested.
01 / OUR PRACTICE
Built by us. Tested in our work.
Research, publishing, sales and development. Our services grow from systems we use, check and improve ourselves.
FROM INPUT TO HUMAN REVIEW
See the input. See the human decision.
For quotation follow-up, the system prepares a reply draft from the quote and its status. A person reviews it before sending. The public demo makes that flow visible.
The demo does not run new AI inference or send email. It is not evidence of client results.
Inspect the input and outputResearch → publish
Primary-source research, bilingual publishing, and checks of the published pages.
Distribute → verify
Adapt articles for social channels and record success, failure, and pending checks.
Collect → decide
Combine replies, delivery failures, and website responses to inform the next sales decision.
HUMAN DIRECTION / AI TEAM
People set direction. AI teams move the work forward.
For example, when exploring a new service.
Human decisionsCoordinationResearchBuild
People set the direction
Define the idea to explore and the constraints to respect.
AI researches and prepares
Share research and prototyping tasks, then assemble evidence and options.
Key decisions return to people
Review the options and open questions, then choose a direction.
Build, check and deliver
Work within the agreed scope and record the outcome and next checks.
Explore the next possibility
Use the results to decide what is worth trying next.
A conceptual illustration of how people and AI can work together, not a client case study. Scope and decision points are designed for each workflow.
IN OUR OWN BUSINESS
What this looks like in our own work.
From research and publishing to checking the response. Selected workflows used in our own operations, with recorded execution evidence.
Used in our own operations
From articles to social posts
We adapt our articles into social copy and images, check publication results, and record failures and pending confirmations.
Carry a useful insight through to distribution.
Used in our own operations
Automated sales reporting
We collect email replies, delivery failures, and website responses. Reports update automatically to inform our next sales decisions.
Bring responses together to decide what comes next.
Examples from A&A’s own operations. These workflows include human decisions and review, and inform the support we provide.
Workflow examples from our own operations
Recorded execution
From research to bilingual publication
- Research primary sources
- Draft in both languages
- Check and publish
- Verify the published pages
A recorded A&A publishing run, including checks of the published Japanese and English editions.
Recorded execution
From an article to publication checks
- Start with an A&A article
- Adapt for social channels
- Publish
- Record success, failure, or pending
An example from our own distribution workflow. Confirmed posts are kept separate from those still awaiting verification.
Recorded execution
From sales responses to the next decision
- Check replies and delivery failures
- Collect website responses
- Update the sales report
- Decide the next action
An example of combining replies, delivery failures, and website responses in our own operations. Reporting activity is not a claim of increased revenue.
BUILT WITH AI. RUN WITH AI.
We build with AI.
We work with AI.
Strategy, sales, creative work, and engineering. We combine specialist agents and reusable workflows, testing and improving them in our own business.
That practice informs your strategy and delivery.
Conceptual illustration of connected AI teams
- Specialist agents
- OVER50AI AGENTS
- Reusable skills
- OVER200WORK SKILLS
- Business functions
- ACROSS08BUSINESS FUNCTIONS
Verified September 2026 · How we count
We counted 54 agent definitions and 229 skill definitions in our development and business environments, excluding personal skills and duplicates. Agents include environment configuration and project-local role definitions, not concurrent running processes or employees. The eight functions group verified capabilities by business purpose.
OUR AI TEAM
AI capabilities for each part of the business.
01
Business strategy
Business options, pricing, and priorities
02
Sales & proposals
Discovery, proposals, and follow-up
03
Marketing
Social media, SEO, and AI search
04
Creative & design
Articles, imagery, and web design
05
Engineering & quality
Design, development, tests, and reviews
06
Research & knowledge
Primary research and reusable knowledge
07
Finance & administration
Invoice and receipt organization
08
Meetings & operations
Meeting preparation, records, and email
OPEN SOURCE
Public development resources
dwhy
A CLI for recording and searching what a project decided and why. Published under the MIT license.
LLM Wiki Bundle Design
Knowledge-base design and templates for AI retrieval. These are design materials, not a finished business system. MIT license.
Our illustrated introduction explains the service separately from the operational records above. Watch the introduction (Japanese audio)
LEARNING PROJECT
LLCC — learn to build with AI
Let’s Learn Claude Code is a learning project for non-engineers building systems with AI. We are preparing it for open-source learning. Public learning materials and enrollment are not yet available.
Preparing for public release — Automate & Augment / Kazuya Hibara
LET’S TALK
Let’s start with your business.
Use the button to book a time that suits you, right away. You do not need AI expertise or a ready-made plan.
Book a free assessment
Free, online, pick a time that suits you