Pluris - Production

← Case Studies

AI Exam Question Generation for Legal Professional Training

GenFutures Lab fine-tuned OpenAI models on Azure for BPP law school's SQE exam bank, cutting review time 75%.

75% Reduction in Review Time

Across SQE exam question production

4–8 weeks

Implementation Time

Mel Moeller

Co-founder @

GenFutures Lab

The Challenge

BPP University needed to rapidly expand its bank of multiple-choice questions for the UK Solicitors Qualifying Examination (SQE). Crafting each legally compliant question took up to an hour because wrong-answer distractors must be plausible and precise. With growing student demand, the law school lacked capacity to scale production through manual effort alone.

What They Built

GenFutures Lab deployed a pair-prompting sprint methodology pairing AI engineers with BPP legal SMEs to fine-tune OpenAI LLMs on a structured legal knowledge base, producing SQE-compliant multiple-choice questions at scale — shifting SMEs from question drafting to review and approval.

GenFutures Lab used a rapid create-test-learn sprint methodology, pairing an AI engineer directly with BPP legal SMEs. The team fine-tuned OpenAI LLMs on Microsoft Azure on a structured legal knowledge base. Iterative feedback loops refined both prompt design and model output across sprint cycles until question accuracy met exam-board standards. SMEs shifted from drafting questions (up to an hour each) to reviewing AI-generated questions (15 minutes each), resulting in a 75% reduction in review time and approximately £10,000 in monthly savings.

AI Role

SMEs went from reviewing 10 questions per day to 40. Monthly output reached 500 high-quality MCQs.

AI Model

Azure OpenAI

Infrastructure

• OpenAI LLMs (fine-tuned on legal knowledge base) • Microsoft Azure (hosting) • SQE question handbook (training corpus)

Integration Points

• SQE handbook → LLM fine-tuning • OpenAI LLMs → Azure-hosted generation endpoint • SME review → iterative prompt refinement

Impact

4x SME Throughput & 500 Questions/Month

SMEs went from reviewing 10 questions per day to 40. Monthly output reached 500 high-quality MCQs.

£10,000 Monthly Cost Savings

The efficiency gains translated to £10,000 per month in cost savings within four months, delivering clear quantifiable ROI.

75% Reduction in Review Time

Time to produce each exam question dropped from 1 hour to 15 minutes, freeing SME capacity for higher-value review and editorial work.

Technology Utilized

Cost

No items found.

Large educational institutions and professional certification bodies in early-to-mid AI adoption that need a structured approach to high-volume, accuracy-sensitive content generation.

Repeatability:

Mel Moeller

Co-founder @

GenFutures Lab

Former Sky, BBC & HP, is an AI thought leader who helps organizations rapidly adopt AI, drive measurable ROI, scale innovation, and foster AI literacy across diverse industries.

Connect with the Expert

Industry:

Education & EdTech

Business Function:

Operations

Product & Engineering

Company Size:

1,001-5,000

Project Cost:

Cost not disclosed

Ownership:

Nonprofit / Institutional

Organization Type:

Nonprofit/Institution

AI Pattern:

Generative Design & Content

Process Automation (RPA + AI)

Value Type:

No items found.

AI Model:

Related Case Studies

AI Knowledge Base for RFP Automation in FinTech Companies

Siloed RFP Process Cut from 38+ Hours to 8–12 Hours with AI Knowledge Base

The Grin Labs built a ChatGPT knowledge base and Custom GPTs for a global FinTech company, cutting RFP response time by 75%+ from 38+ hours to 8–12 hours per response.

RFP Response Time Cut by 75%+

From 38+ hours to 8–12 hours per

Sales & Revenue

,

Operations

,

< 4 weeks

Financial Services

,

Knowledge Management & Search (RAG)

,

Generative Design & Content

,

Cost Reduction

,

This is some text inside of a div block.

Carl Miller

Head of AI • The GRIN Labs

Automated LinkedIn Content Pipeline for SMB Professional Services

A Full Week of LinkedIn Content in 30 Minutes: No-Code AI Pipeline Frees SMB Owners from Daily Social Media Work

Panjutorials GmbH built a Make.com pipeline connecting Airtable, OpenAI, and Perplexity AI to generate and publish a week of LinkedIn content in under 30 minutes for SMB owners.

Full Week of Content in Under 30 Minutes

Versus daily manual research and

Marketing

,

6–12 months

Professional Services

,

Generative Design & Content

,

Process Automation (RPA + AI)

,

No items found.

This is some text inside of a div block.

Denis Panjuta

CEO • Panjutorials GmbH

AI-Powered Go-to-Market Optimization for Financial Services

Stalled Product Launch → 200% Lead Response Lift, 12% Sales Lift, 40% Productivity Gain

Human Pilots AI used Claude, ChatGPT, and Gemini in structured AI sprint pods to reposition a retail payments product, delivering 200% lead response lift and 12% sales increase.

200% Lead Response Rate Lift

During AI-assisted outreach pilot

Marketing

,

Sales & Revenue

,

Executive & Strategy

,

4–8 weeks

Financial Services

,

AI Workforce Enablement

,

Generative Design & Content

,

No items found.

This is some text inside of a div block.

Sean Wood

Founder • Human Pilots AI