Prologue · Vision

I'm Ho Daniel Khuu data scientist & and full-stack developer

View Projects📄 Download CV
Scroll to follow the next chapter
Chapter 01 · Origins

Building bridges between math, machines, and meaningful stories

I partner with organisations to translate messy datasets into decisions people trust. From Oslo to remote teams worldwide, I deliver products where analytical depth meets narrative clarity. I also run Xaku, a design & technology studio focused on high-impact digital experiences.

Based in

Oslo, Norway · working globally

Focus

Data science, full-stack development, narrative-led strategy

CompanyXaku · Design & teknologi studioBesøk nettsiden ↗
Ho Daniel

Follow the milestones that shaped how I build—each chapter adding new tools, perspectives, and empathy for the teams I support.

2020
Analytics Foundations

Google Data Analytics Certificate

Completed Google's professional certificate to formalise core analytics practices—SQL, dashboards, and stakeholder-ready storytelling.

2021
Community Leadership

Activity lead & leadership training

Led multi-sport programmes at Tøyen Sportsklubb while completing 150+ hours of leadership training focused on facilitation and coaching.

2024
Product Innovation

Silver Beats biometric music venture

Co-founded a student company building a Spotify-integrated coaching app—driving research, prototyping, and investor-ready presentations.

2025
Data Science Graduate

Bachelor of Data Science, Høyskolen Kristiania

Earned top marks across machine learning, predictive modelling, and communication—graduating with an A on the bachelor thesis.

2025
Full-Stack Delivery

App developer for VitusApotek wellbeing platform

Designed and shipped a React Native + Node.js solution backed by Azure that energised 3k+ employees with nudges, analytics, and GDPR-safe data flows.

Chapter 03 · Impact

Projects told as miniature case studies

Each card captures a pivotal moment—what needed to change, how I approached the challenge, and the results that followed.

Air Quality Prediction
Data Scientist & ML Engineer

Air Quality Prediction

Forecasting pollution events with ensemble models, real-time pipelines, and actionable dashboards for municipalities.

ChallengeMunicipal planners needed to anticipate pollution events fast enough to warn residents and adjust traffic flows.

ApproachBuilt an ensemble forecasting pipeline that cleanses multi-source sensor data, engineers weather features, and serves predictions through a containerised API.

OutcomeDelivered 94% accurate forecasts powering daily dashboards and automated alerts across three city districts.

PythonTensorFlowScikit-learnDocker
Read case study ↗
Trading Journal
Full-Stack Developer

Trading Journal

A personal trading journal with emotion tracking, analytics dashboard, and live broker API integration — built on Supabase.

ChallengeTraders often let emotions drive decisions without a structured way to log, reflect, and spot behavioural patterns across sessions.

ApproachBuilt a full-stack journal app with Supabase as the backend, covering trade logging, emotional state tagging per entry, a real-time analytics dashboard, and OAuth-secured broker API connections for automated data import.

OutcomeDelivered a unified workspace where traders can review performance, identify emotional triggers, and connect directly to trading platforms for seamless data sync.

Next.jsTypeScriptSupabasePostgreSQL
Read case study ↗
Xaku Intake Agent
AI Agent Engineer & Backend Developer

Xaku Intake Agent

A client-intake agent for Xaku that qualifies leads and answers pricing questions, but can never confirm a price, deadline, or scope on its own — every commitment is routed through a deterministic, code-level approval gate.

ChallengeA solo studio's intake needed to qualify leads and answer pricing questions while Daniel is unavailable — without an LLM ever being able to quietly promise a price, deadline, or scope it shouldn't.

ApproachBuilt a FastAPI backend around Claude's tool-calling loop with six single-purpose tools (qualify, answer, scope, book, escalate, propose-commitment) and a guardrail layer that intercepts every tool call in code before it can take effect — there is deliberately no tool that confirms a commitment, only one that queues it for human approval. Every reasoning step, tool call, and guardrail decision is written to an append-only trace, visible live on a self-contained showcase page with its own chat, session-trace, and password-gated admin queue.

OutcomeShipped with a 141-test suite covering the full guardrail matrix (price floors, minimum lead time, out-of-scope and always-escalate categories) plus scripted end-to-end and adversarial/prompt-injection cases — all runnable without a live API key via a pluggable scripted model client.

PythonFastAPIAnthropic Claude APIPydantic
Read case study ↗
View All Projects