Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Career AI: Kenyan Student Pathways
Learn how we built a conversational system that uses NLP, MBTI and interest profiling to match Kenyan Grade 9 students with CBC pathways and career resources.
Career AI is an intelligent guidance platform that helps Kenyan students in the Competency-Based Curriculum (CBC) system select their Grade 9 career pathways using psychometric data and conversational intelligence.
In this demo, I’ll show how we built the engine, from natural language processing that interprets students’ questions, to personality mapping (MBTI + Interest Profiler) and adaptive advice generation.
I’ll walk through:
How student inputs are tokenized and contextually mapped to personality traits and local career clusters.
How the model evaluates compatibility between a student’s traits, interests, and Kenya’s CBC pathways.
How we built the resource center integration that surfaces localized occupational insights and KNEC-aligned content.
Gamified AI assesses MBTI types via scenario-based questions for career matching.
- GPT-4GPT-4 is OpenAI’s large multimodal model: it processes both text and image inputs, delivering human-level performance on complex professional and academic benchmarks.This is OpenAI’s latest milestone in scaling deep learning: a large multimodal model accepting both text and image inputs. It demonstrates a significant capability leap over its predecessor, scoring in the top 10% on a simulated bar exam (GPT-3.5 scored in the bottom 10%). The model handles nuanced instructions and long-form content, supporting context windows up to 32,768 tokens (32K model). This capacity allows processing up to 25,000 words in a single, complex prompt. GPT-4 is engineered for enhanced reliability, steerability, and advanced reasoning across diverse tasks.
- LangChainThe open-source framework for building and deploying reliable, data-aware Large Language Model (LLM) applications.LangChain is the essential framework for engineering LLM-powered applications: it simplifies connecting models (like GPT-4 or Claude) to external data, computation, and APIs. The platform provides a modular set of components—Chains, Agents, Tools, and Memory—allowing developers to quickly build complex workflows like Retrieval-Augmented Generation (RAG) pipelines and sophisticated conversational agents. Its Python and JavaScript libraries, combined with LangChain Expression Language (LCEL), offer a standardized interface for rapid prototyping and moving applications to production with confidence.
- FastAPIFastAPI is a modern, high-performance Python web framework for building APIs with automatic OpenAPI documentation.FastAPI is a robust, high-speed Python web framework: it is built on Starlette (for async capabilities) and Pydantic (for data validation and serialization). Leveraging standard Python 3.8+ type hints, the framework automatically generates interactive API documentation (Swagger UI/ReDoc) and enforces data validation, effectively reducing developer-induced errors by an estimated 40%. This architecture delivers performance on par with Node.js and Go, significantly increasing feature development speed (up to 300% faster). It is production-ready, fully supporting OpenAPI and JSON Schema standards for all API specifications.
- RenderRender is the unified cloud platform: build, deploy, and scale your full-stack applications and services with automatic zero-downtime deploys and managed infrastructure.Render delivers accessible, reliable cloud infrastructure, simplifying deployment for all service types: Web Services, Static Sites, Background Workers, and managed databases like Render Postgres. The platform supports native runtimes (Node.js, Python, Ruby) or Docker, offering features like load-based autoscaling and Private Networking. Teams leverage Infrastructure as Code (IaC) via Blueprints to define and replicate entire environments quickly. With over $155M in funding, Render is engineered for speed and reliability, supporting everything from prototypes to applications with hundreds of services at global scale.
- PostgreSQLPostgreSQL (Postgres): The world's most advanced, open-source object-relational database (ORDBMS), built for reliability and extensibility.PostgreSQL is the premier open-source ORDBMS, proven over 35+ years of active development. It adheres strictly to ACID properties (Atomicity, Consistency, Isolation, Durability), ensuring data integrity for mission-critical workloads. Key features include robust SQL compliance, Multi-Version Concurrency Control (MVCC), and superior extensibility (e.g., custom data types, functions in multiple languages). Advanced capabilities like native JSON/JSONB support and the PostGIS extension (geospatial data) make it a powerful, flexible choice for complex enterprise applications.
Related projects
Educational AI
Toronto
Learn how we built a code‑annotation model and a documentation generator, the challenges faced, and insights for improving…
IncluLearn Global: An AI-Powered School & Health Recommendation Engine with Explainable Suggestions
Nairobi
A step‑by‑step demo of building a school and health recommendation engine using Supabase, vector embeddings, GPT explanations, and…
AI Powered Personal Knowledge Management Tool
Nairobi
See how to collect LinkedIn, Twitter, Instagram, Reddit posts, use AI to extract key details, and store them…
Automated Sales Assistant That's a Helpful "know-it-all"
Nairobi
Learn how to build a semantic‑search‑driven sales assistant using open‑source tools, local LLMs, and scalable vector similarity for…
Building a Conversational AI Using ElevenLabs, React, and FastAPI
Nairobi
Learn to integrate ElevenLabs TTS with React and FastAPI to build a real-time conversational AI, covering voice synthesis,…
Rafiki AI Tutor
Nairobi
Demonstration of Rafiki, an AI tutor that uses SMS and voice calls on basic Kenyan phones to deliver…