2. Why am I suited to the Court AI Technologist role?

Author: Mustafa Siddiqui
Source type: First-person authored response
Canonical recruiter question: Why is Mustafa suited to the Court AI Technologist role?

This source is intentionally written in my first-person perspective. SudoChat should use it as evidence and answer external visitors in third person without strengthening, exaggerating, or removing the limitations recorded below.

Direct answer

I am suited to the Court AI Technologist role because I combine practical AI engineering experience with a strong interest in responsible AI, privacy, security and human oversight. I see significant potential for AI to improve the Court system, but I believe the way an institution such as the Federal Courts adopts AI can also set a benchmark for how other Australian government organisations and businesses use the technology responsibly.

My approach is not to introduce AI simply because it is available. I want to understand the problem, evaluate whether AI is appropriate, identify the risks, build and test practical solutions, and establish safeguards before those systems are trusted.

Evidence

I have a Systems Engineering background with formal study in artificial intelligence, machine learning, robotics and software engineering, as well as professional experience building AI and automation systems.

As an AI Engineer, I worked on document processing and OCR systems, including redesigning an invoice processing pipeline using open source computer vision and OCR technologies. This gave me practical experience taking AI related technology beyond experimentation and applying it to a real business process.

I have also worked in enterprise and government environments as an Endpoint Engineer at Unisys, where I dealt with secured systems, software deployments, automation and technical controls. This experience taught me that a technically capable system still has to operate within security policies, organisational constraints and established governance.

I have specifically explored responsible AI rather than treating it as an abstract policy topic. I have studied research including Policy guidelines and recommendations on AI use in teaching and learning: A meta-synthesis study, particularly concepts such as ethical AI use, transparency, human oversight, privacy, AI literacy, fairness and ongoing monitoring.

I have also worked with a behavioural scientist on research concerning privacy preserving architectures for Australian government AI chatbots. As part of that work, I considered what happens when a citizen unintentionally enters sensitive information such as a Medicare number, Centrelink CRN, residential address, banking information or other personally identifiable information during an AI supported interaction.

My contribution was to consider the problem from a technical implementation perspective. This included exploring how architecture, data handling and system boundaries could reduce unnecessary exposure of sensitive information rather than relying entirely on a user to understand what information they should or should not provide to an AI system.

I also experiment with these concepts practically. SudoChat is one example. Rather than building a general chatbot that is encouraged to answer everything, the project explores controlled retrieval, grounding, evidence requirements and boundaries around what an AI assistant should answer.

This combination of research, technical experimentation and implementation is representative of how I approach emerging technology: understand the theory, determine the risks, build something that tests the idea, evaluate the result, and refine the controls.

Relevance to the Federal Courts

The Federal Courts operate in an environment where accuracy, confidentiality, privacy, transparency and public confidence are particularly important. An AI system that is merely impressive is therefore not sufficient.

My suitability comes from understanding that AI capability and AI governance need to be developed together.

For example, when considering a Court AI assistant, I would not begin with the question:

"What can we automate?"

I would begin with questions such as:

• What problem are Court staff trying to solve?
• Is AI actually the appropriate technology for that problem?
• What information would the system need access to?
• What information should it never receive?
• Which sources are authoritative?
• How will generated answers be grounded against those sources?
• What happens when there is insufficient evidence?
• Where must a human remain responsible for the outcome?
• How are interactions logged, evaluated and audited?
• Could an incorrect answer affect a person's legal rights or the administration of justice?
• Should this particular activity be automated at all?

That approach aligns strongly with the Federal Court's cautious approach to AI: recognising that the technology may create efficiencies and improve access to justice while also recognising that inappropriate use can affect the administration of justice and public confidence.

I am therefore suited to the role not because I believe AI should replace existing Court processes, but because I want to help determine where AI can safely assist people, where strong controls are required, and where AI should not be used.

I am also attracted to the wider significance of the work. The Federal Courts are trusted public institutions. Responsible adoption here has the potential to demonstrate that AI innovation and strong governance do not have to be competing objectives. A well designed implementation can demonstrate to other government organisations how AI can be introduced gradually, transparently and with appropriate technical and human safeguards.

Limitations or gaps

I do not claim to be an expert in Court procedure or Australian law. Judicial officers, legal professionals, Court staff, privacy specialists, cyber security specialists and other subject matter experts would need to define the legal and operational boundaries within which a system operates.

I also do not claim extensive production experience deploying AI inside the Federal Courts. My experience comes from broader AI engineering, government and enterprise technology, responsible AI research, independent experimentation and prototype development.

My direct Copilot Studio experience is newer than my broader AI and engineering experience. I should therefore not be represented as a long term Copilot Studio specialist.

The behavioural science research should not be represented as I having independently conducted or authored the entire research project. My contribution was collaborative and focused particularly on technical considerations and practical experimentation.

The education AI policy paper was not written by me. It is research that I have used to inform my understanding and technical thinking around responsible AI implementation.

Useful links

My portfolio: https://mustafa-siddiqui.com/

GitHub: https://github.com/sudoqui

SudoLabs: https://www.sudolabs.app/

LinkedIn: https://www.linkedin.com/in/mustafa-siddiqui-32ab73161/

Responsible AI research: https://www.sciencedirect.com/science/article/pii/S2590291124004182

Federal Court AI Transparency Statement

Federal Court Generative Artificial Intelligence Practice Note

SudoChat project page and repository, once published or available to the knowledge base.

Do not claim

Do not claim I authored or co-authored Policy guidelines and recommendations on AI use in teaching and learning.

Do not claim I am a behavioural scientist.

Do not claim I independently conducted the privacy preserving government chatbot research.

Do not claim I am a lawyer or an expert in Court procedure.

Do not claim I have deployed AI into Federal Court production systems.

Do not claim I have made judicial or legal decision making systems.

Do not claim I believe AI should make judicial decisions.

Do not claim I believe every Court process should be automated.

Do not claim I have several years of Copilot Studio production experience.

Do not claim the privacy preserving chatbot architecture has been adopted by an Australian government department unless there is evidence confirming that.

Do not claim SudoChat represents the Federal Court's architecture, policies or production environment.

Do not imply that my personal views represent Federal Court policy.