Driving Workflow Transformation Through AI-First Development at TEX for Schools
How AI-first development workflows boosted team productivity by 5x
Overview
TEX for Schools is a SaaS platform designed to empower school counselors in Australia by simplifying how they support students with opportunities, mentorship, career pathways, and alumni connections. As the product lead, Kunal recognized early on that traditional development cycles would not suffice for the speed and adaptability required in the education sector. With a lean team and ambitious goals, TEX needed a radical shift in how software was built and delivered.
Challenge
TEX for Schools faced several critical challenges that threatened their ability to deliver value quickly to school counselors: small team with limited engineering bandwidth, diverse feature requirements from multiple user roles (counselors, students, alumni), the need for rapid iteration while maintaining high product quality, counselor-facing UI required a minimal learning curve and high responsiveness, and fragmented knowledge management and repetitive dev tasks slowed velocity.
Solution
Kunal introduced an AI-first development workflow across the TEX engineering stack, with the goal of turning the team into designers of intent rather than manual executors of code.
Impact ✦
| Metric | Before (Baseline) | After AI-First Integration |
|---|---|---|
| Feature delivery time | 1-2 weeks per feature | 2-3 days per feature |
| Code review turnaround | 1-2 days | Same day (via AI pre-review) |
| Bugs from new feature releases | Moderate (20% of releases) | Reduced by >50% |
| Team dependency on senior dev | High | Significantly lower |
| Onboarding time for new devs | ~2 weeks | <5 days |
Cursor IDE for Development Acceleration
Adopted Cursor as the primary IDE for the dev team, leveraging its native GPT-4 integration to automate boilerplate generation, refactor legacy code, and generate unit tests with near-zero friction.
The team was able to focus on higher-level architectural decisions while AI handled repetitive coding tasks.
This transformation allowed junior developers to produce senior-level code quality through AI assistance.
Prompt-Driven Architecture Design
Used prompt engineering to rapidly iterate and test system design ideas, database schemas (Supabase RLS policies), and API contracts before writing a line of code.
This replaced lengthy planning meetings with async prototype generation and validation.
The team could explore multiple architectural approaches quickly, leading to better informed decisions.
MCP and Background Agents
Leveraged background agents within Cursor to automate code scaffolding, linting, and test writing — enabling junior developers to produce production-grade code faster and with more confidence.
Automated quality checks caught issues before they reached code review, improving overall code quality.
This reduced the mentoring burden on senior developers while maintaining high standards.
Playbooks for Consistency
Developed a set of reusable AI prompts and project-specific playbooks (e.g., "Add a feature for counselor role", "Secure an endpoint using Supabase RLS") to ensure consistency and reduce decision fatigue.
These playbooks became institutional knowledge that could be easily shared and updated.
New team members could immediately follow established patterns without extensive onboarding.
Knowledge Codification
Institutionalized AI-augmented documentation. Architectural decisions, RLS logic, and onboarding flows were written once, reviewed by AI, and updated through AI-assisted edits — keeping team knowledge up to date without relying on memory or tribal knowledge.
Documentation remained current and comprehensive without manual maintenance overhead.
This created a self-reinforcing system where knowledge improved over time.
Cultural Transformation
Freed up senior engineering time to focus on product thinking instead of code firefighting.
Created a culture of async-first thinking — AI became the first sounding board before Slack threads or meetings.
Enabled more rapid experiments with UX, counselor flows, and student onboarding without major resource overhead.
The team developed confidence to tackle larger, more ambitious features knowing AI would support the implementation.
Conclusion
By embedding AI deeply into the development lifecycle — not just as a helper, but as a core operating principle — Kunal helped transform how TEX for Schools was built. The AI-first approach didn't just boost speed; it reduced burnout, raised code quality, and created an adaptable development system that scales with product ambition. This transformation reflects a broader truth: for lean teams solving complex problems, AI-first development is not a luxury. It's leverage.
Ready to Transform Your Development Workflow?
Book an AI Build Session and get expert guidance to implement similar transformations in your team.
Book Your Session