SAMUEL
GETACHEW.

Building intelligent systems, creating useful products.

I am Samuel, a software engineer and founder. I turn complex problems into useful products, applying AI where it makes a real difference.

Scroll

01 / A little context

I turn complex problems into useful, intelligent products. From autonomous agents to knowledge systems, I bring an engineer’s precision and a founder’s perspective to everything I build.

02 / Selected work

Ideas, made real.

From venture-building agents to
knowledge that finds its way off the shelf.

03 / Experience & toolkit

Built in the work.

  1. Aug 2025 — Present

    Yuba Labs Ltd

    Co-founder & AI Engineer

    Kigali, Rwanda

    Building agentic products for venture creation, founder recruitment, and investment research. Leading development, workflow evaluation, and platform performance across Yuba, Apptus, and YubaVC.

  2. Nov 2024 — Aug 2025

    Dossie Technologies

    Co-founder & AI Engineer

    Addis Ababa, Ethiopia

    Built OCR and RAG knowledge systems, including ingestion of 15,000 university research papers. Validated a medical-record digitization workflow with 167 healthcare professionals across 147 facilities.

  3. Jun 2022 — Apr 2023

    AgriStar

    Co-founder & Software Engineer

    Addis Ababa, Ethiopia

    Connected smallholder farmers with markets and agricultural advisory services. The venture placed third in the AYuTe Africa Challenge.

  4. Oct 2022 — Feb 2023

    Safe Transport Technology

    Software Engineering Intern

    Addis Ababa, Ethiopia

    Built a Telegram-based research tool using Telepathy and structured more than 200,000 target-market data points to support market analysis and outreach.

Across the workWhat I Build With3 areas

Curiosity drives the learning. Real problems shape the toolkit.

Intelligent systems

  • Agentic workflows
  • RAG
  • LangChain
  • LangGraph
  • Azure OpenAI
  • PageIndex
  • Qdrant
  • pgvector
  • OCR
  • LLM evaluations
  • Data pipelines

Product engineering

  • Python
  • TypeScript
  • JavaScript
  • SQL
  • React
  • Next.js
  • FastAPI
  • PostgreSQL
  • Supabase
  • Three.js
  • WebGL

Architecture & delivery

  • Product discovery
  • Rapid prototyping
  • Multi-tenant Systems
  • Authentication & RBAC
  • Microsoft Azure
  • Redis
  • Celery
  • Locust
  • Performance testing

04 / Blogs

Thinking out loud.

Notes from building AI systems,
and the questions they leave behind.

A silver sphere enclosed by delicate wireframe boundaries, representing web access and permission.

AI Engineering / System Design

What Building AI Agents Taught Me About Web Access and Permission

Lessons from building AI workflows at Yuba: how web retrieval differs from permission, where quality checks stop, and what explicit access boundaries require.

6 min readRead article
A restrained wireframe staffing grid with an open position and a silver reserve sphere, representing forecast uncertainty and qualified backup capacity.

Applied Machine Learning / Aviation Operations

When Airline Crew Do Not Return: Forecasting the Staffing Risk

An engineering proposal for airline crew nonreturn: forecast staffing gaps, plan qualified reserves, and assess the limits of machine learning and LLMs.

14 min readRead article
Floating information sheets linked by a fine orbital path, representing the flow and value of information.

Information Economics / AI Product Design

How AI Agents Could Change the Economics of Online Information

How AI agents could affect publisher revenue, why attribution is not a payment mechanism, and how to compare licensing models using explicit assumptions.

6 min readRead article

Engineering, ideas, and lessons from the work.

More blogs 07

05 / Curiosity, off the clock

Hobby Lab.

A place for the things I want to make.
Robotics, flying machines, and playful experiments.

Selected from the build queue06 / 26

Ideas collected from other makers.
These are inspirations for future builds.

More projects 26