About Iras Lab
Advisory Work That Sits Beside the Engineer, Not Above Them
Iras Lab is a small advisory practice based in George Town, Penang. We help QC teams understand where AI reading tools can help, and where they should stay out of the way.
← Back to HomeOur Story
How Iras Lab Came to Be
Iras Lab was founded after a series of conversations with QC managers at manufacturing facilities across Penang who were receiving a lot of vendor pressure to "adopt AI" — without clear guidance on what that actually meant for their ISO 9001 documentation trails or their Halal audit expectations.
The practice was built on one observation: most AI integration advice is written for software teams, not for engineers who manage physical inspection records, non-conformance logs, and corrective action procedures. The vocabulary is different. The stakes for getting it wrong are different.
We started small — a single engagement reading inspection records for a food processing facility in Bayan Lepas. That work shaped everything that came after. Our engagements are still deliberately limited in scope, because the best AI integrations in quality control are the ones that earn trust incrementally.
Our Mission
What We Are Here to Do
Help QC leaders make considered decisions about where AI reading tools belong in their workflows — without pressure to move faster than their teams are comfortable with.
Ensure that any AI integration we recommend is compatible with the standards your facility operates under — ISO 9001, JAKIM Halal Manual, customer audit requirements.
Leave your QC team with clearer documentation of how AI is being used, so they can explain it confidently during audits and management reviews.
Preserve engineer authority at every step. AI surfaces information. Your inspectors decide what it means and what to do about it.
The Team
The People Behind the Work
Nadia Azman
Principal Advisor
Fourteen years in quality systems across food manufacturing and medical device supply chains in Penang and Kedah. Led the practice's first engagements with LIMS-integrated AI reading tools.
Rafi Zulkifli
QMS Integration Lead
Former QC engineer with a background in ISO 9001 internal auditing. Manages the read-only integration configuration for all Pilot engagements and inspector training.
Siti Cheng
Standards & Compliance
Specialist in Halal certification documentation and CAPA process design. Reviews all briefs for standards alignment before delivery to clients.
How We Work
Our Professional Standards
Read-Only Access Policy
We access client systems in read-only mode only. No changes are written to your QMS or LIMS without explicit instruction from your QC team. This is a non-negotiable operating principle.
ISO 9001 Awareness
All recommendations are reviewed against ISO 9001 clause requirements before delivery. We identify any areas where AI tool use would need documentation to maintain audit trails.
Halal Manual Alignment
For facilities with Halal certification, we cross-reference JAKIM MS 1500 requirements and ensure that AI-assisted record handling does not introduce compliance gaps.
Data Handling Protocol
Client inspection records and batch data are used solely for the engagement in scope. We do not retain, aggregate, or use client data beyond the agreed deliverable. Data is returned or deleted on engagement close.
Written Deliverables Only
Every engagement closes with a written brief — plain language, addressed to your head of quality. Verbal recommendations without documentation are not our practice.
Engineer-First Principle
We do not design engagements that move decision-making authority from engineers to algorithms. If a recommendation would reduce engineer oversight, we do not make it.
Our Approach
AI in Quality Control: What Iras Lab Actually Does
Quality control in manufacturing — particularly in food, pharmaceutical, and precision component sectors — depends on documentation that holds up to scrutiny. Inspection records, non-conformance reports, corrective action logs, and CAPA memos are not just operational tools. They are the evidence base for your ISO 9001 certification, your Halal status, and your customers' confidence in your process control.
When AI reading tools are introduced without care into this environment, the risks are not technical failures — they are documentation failures. An AI that summarises records inaccurately, or drafts a CAPA memo that omits a root cause step, can create audit discrepancies that take weeks to unravel. This is the problem Iras Lab was built to address.
Our advisory work is narrow by design. We do not build AI systems. We do not implement software. We read your records alongside your team, map where an AI reading assist would reduce administrative load without introducing documentation risk, and write you a clear, jargon-light brief that your QC manager can act on or set aside as they see fit.
The QC leaders we work with in Penang, Kedah, and the Klang Valley tend to be experienced engineers who are open to the idea of AI support but cautious about the implementation. That caution is well-founded, and we share it. The engagements we offer are structured to move at the pace of trust, not the pace of a technology sales cycle.
Work With Us
Ready to Have a Measured Conversation?
We are a small practice and we keep our engagements small. If you'd like to discuss whether a Workflow Reading Review makes sense for your facility, get in touch directly.
Get in Touch