FINALISTTop 16 Finalist2026

The Passion Discovery Engine

IntelliCon '26 Buildathon | AIESEC in SLIIT

A mobile-first career discovery platform where Sri Lankan students explore real study and career paths through AI-guided work simulations and adaptive skill mapping.

Engineering Role

Full-Stack Developer

Sprint Duration

Three weeks (20 September-11 October 2026)

Team Configuration

HungryBeast (4 members)

Outcome Status

Finalist

Career DiscoveryAI SimulationsEducationSri LankaIntelliCon 2026

01 // PROBLEM DOSSIER

The Challenge & Core Problem

High-Level Briefing

Sri Lankan students make high-impact A/L stream, degree, and career choices with little structured, hands-on exposure to what those paths involve. The submission focused first on A/L completers waiting to enter university.

Formal Technical Problem Statement

"Help students make more informed career and study choices by letting them experience realistic work scenarios, reflect on their decisions, and discover skills to develop."

Hard Timebox & Rules

Three-week Buildathon, 20 September-11 October 2026

The initial product scope focused on A/L completers in the post-exam gap year.
The proposal planned three pilot simulation areas: IT/Product, Management, and Bio Science.
The platform was planned as mobile-first with English and Sinhala support.

Open Risks & Pressure Factors

Whether students would complete multi-day simulations remained to be tested with beta users.
Tuition-centre distribution and corporate sponsorship interest were still unvalidated.

02 // LEADERSHIP & ARCHITECTURE

Leadership & Personal Contribution

Full-Stack Development

Contributed as the full-stack developer on the four-member HungryBeast team.

Worked on a submission that combined AI-guided career simulations with adaptive skill mapping and peer cohorts.

Real-Time Architectural Dilemmas Under Clock Pressure

DILEMMA #1

How could the first version stay focused while serving students at several education decision points?

➔ Action Taken: The initial product scope targeted A/L completers in the gap year, with O/L and current A/L students treated as future expansion segments.

Engineering Rationale: The report identifies segment dilution as a risk and prioritizes validation with one clear initial audience.

03 // SYSTEM ARCHITECTURE

The Technical Solution

The concept pairs conversational AI career simulations with skill mapping based on students’ work and decisions, then recommends focused learning and peer support.

SYSTEM TECHNOLOGY STACK

Next.jsReactTailwind CSSGemini 2.0 FlashGroqVercel AI SDKNode.jsExpressPostgreSQLSupabase

Data Flow & Architecture

A mobile-first web platform was planned with an AI simulation layer using Gemini 2.0 Flash and Groq through the Vercel AI SDK, a Node.js and Express backend, and Supabase for PostgreSQL, authentication, and real-time peer cohorts.

Critical Data Pipelines

[1]Students work through role-play scenarios based on real tasks in selected careers.

[2]The system evaluates demonstrated strengths and gaps from their reasoning and decisions.

[3]Personalized micro-learning recommendations target identified skill gaps.

[4]Local peer cohorts provide shared support and accountability.

04 // RETROSPECTIVE

The Retrospective: What Worked & What Needs Work

What Worked

✓The Gate 1 report documented primary research with 28 survey responses and five interviews.
✓The team selected a clear first audience and connected the proposed simulations to student research.

What Still Needs Work

✕The report notes that actual completion of multi-day simulations had not yet been tested.
✕Tuition-centre partnerships and corporate sponsorship interest still needed validation.

Hard-Won Engineering Takeaways

01.Test real user behaviour rather than relying only on stated interest.

02.Keep the first product focused on one audience before expanding to adjacent segments.

03.Use beta feedback to improve simulation quality and measure completion.

05 // DEEP-DIVE SPECIFICATION

Research-Led Career Simulations

A focused first step for students in the university-entry gap year

The proposal uses early research to define its beachhead audience, then centers the product around experiential simulations rather than static career descriptions.

Problem research

28 survey responses and 5 interviews

Primary research documented in the Gate 1 submission.

Initial audience

A/L completers awaiting university

The MVP focuses on students in the post-exam gap year.

Pilot career areas

IT/Product, Management, Bio Science

Three pilot simulation scenarios were planned for the initial product.

HungryBeast reached the top 16 finalists at IntelliCon '26 Buildathon.