A studio for expensive assumptions
We notice the problems people have learned to live with, understand why they persist, and build systems that make them easier to solve.
PRETORIA / SOUTH AFRICA
25.7479° S / 28.2293° E
01 / The studio
We start with what isn't working.
The fix isn't always AI, and it isn't always the same fix twice. We inspect the evidence, test the assumption, build what is useful, and revise when reality disagrees.
02 / The engine
A recurring AiForm approach when information, verification, and trust sit at the centre of a problem—not a claim that every project follows the same four stages.
01
Give messy, real-world information a useful shape.
02
Check claims against evidence, not convenience.
03
Turn evidence into something people can act on.
04
Connect the right buyer, supplier, answer, or fit.
Parse → Verify → Understand → Match
03 / Selected work
Different problems. Same instinct.
Product / Procurement / Live
AIFORM / PROCUREA procurement system for public opportunity discovery, supplier compliance evidence, and readiness signals.
View product ↗SUPPLIER OVERVIEW
Updates, product progress and procurement thinking from AiForm Procure.
Products
Studio projects
Experiments
Kutlwano Tutoring Platform
A naming, identity and tutoring website built around clearer learner journeys and lesson booking.
NYAUTSA SS Trading
A portfolio for a Gauteng building business, organised around completed work and direct enquiries.
Brand Identity & Digital Presence
A visual foundation for a handmade business: identity, palette, typography and social templates.
Construction Operations Platform
An investigation into permits, contractor verification and clearer project documentation.
Case study 01 / AiForm Procure
See the problem, the product decisions, the verified capabilities and what the build is teaching us.
04 / The build index
A growing archive of products, client systems and experiments built by AiForm.
2025
2026
2026
Next
Current Build Index
09 active projects across products, client systems and experiments.
Updated from the AiForm Engine operating ledger.
Compliance workflows for South African businesses.
Not announcedPlant pathology and biotechnology research tools.
Not announcedAcademic infrastructure for laboratories and research groups.
Not announced05 / Principle
A configuration isn't correct until it's tested. A feature isn't fixed because the code changed. It's fixed when the behaviour changes.
06 / The system
Different products. One evolving way of researching, building, verifying and operating them.
The Engine describes how information-driven products make sense of complex inputs. The System describes how the Studio preserves context, coordinates work and improves across projects.
Research, documentation, project context and accumulated learning.
AI-assisted analysis, specialised agents and focused workflows.
Reusable methods, standards and patterns.
Shared integrations, automation and infrastructure connecting the work.
Runtime checks, logs, state and feedback.
The point isn't automation for its own sake. The system reduces repeated work, preserves context and makes it easier to test assumptions instead of carrying them from one project to the next.
Lessons from one build don't disappear when the project ends.
Research, development and verification follow repeatable standards.
Reusable systems let more effort go into the problem instead of rebuilding the machinery around it.
The products change. The system learns.
07 / Founder
AiForm Studio is led by Thabiso Eric Motaung—an academic, researcher and product builder who designs and builds each engagement from research through delivery. That combination keeps the work close to how people learn, reason and get stuck.
Identify → inspect → test → build → observe → revise.
08 / The mark
The mark was noticed before it was designed.
One mark / distinct contexts
AIFORM / PROCURE09 / Journal
Content coming soon.
Research, observations and lessons from what we're studying and building.
Visit the Journal →Issue 00
AFS / Field notes
10 / How we work
Sometimes the answer needs AI. Sometimes it needs better data. Sometimes it needs automation. Sometimes it just needs software designed properly.
Technology serves the problem, not the other way around.
11 / Where next
Documents everywhere. Important claims that are difficult to verify. Two parties deciding whether they can trust one another. Repetitive work that software could handle more clearly.