Keith Sng

Analytics engineer and data analyst.

I build data pipelines, models, and dashboards, and more recently the AI tools that sit alongside them.

About

I’m a Senior BI Analyst (AI) at Monee, where I build data pipelines, dashboards, and AI workflows for the business. Being the bridge is where I work best: understanding what people need, working out how the data can get there, and delivering insights back to them.

Coming from a non-traditional background, I started in Environmental Studies at NUS, which taught me to thread at the intersections of multi-disciplinary problems and communicate findings to different audiences. Wanting to be a conservationist led me to the National Parks Board, where I did applied ecological research and programme management, from estimating national stray animal populations to building partnerships with the community and agencies.

I then moved to Singlife and into data analytics and engineering, building the pipelines and data models that actuarial and business teams relied on daily. Working between the business and the technical side, I turned business needs into reliable, automated data foundations, multiplying the impact of the teams who used them.

As a curious learner, I'm constantly on the lookout for new opportunities to learn and grow. First, with a Specialist Diploma in Data Science at Singapore Polytechnic, and now a Master’s in Computer Science at Georgia Tech, specialising in AI, to strengthen my foundations in machine learning and systems design for production-ready solutions.

If my work interests you, let's connect and chat!

Skills

Used most

Languages
SQL Python R Java C
Data Engineering
Snowflake Airflow Spark SQL Presto Hive AWS S3 Alteryx
Data Visualisation
Qliksense Streamlit Plotly Seaborn Matplotlib
Analytics & ML
Pandas Scikit-learn AWS SageMaker Kedro XGBoost PyTorch TensorFlow MLflow Optuna SciPy
AI Tools
Claude Code Codex Herdr Cursor Topview HeyGen
Development & Collaboration
Git Jira Confluence

Experience

Monee

Senior Business Intelligence Analyst (AI)

Jun 2026 – Present

  • Created a GenAI video marketing pipeline (prompts, storyboard frames, interpolation, voiceover) that shipped 3 campaign videos across Malaysia and Singapore in 3 months and cut MVP turnaround to under 5 manhours. Developing a self-serve playbook so business development teams can produce localised videos at scale.
  • Engineered a traffic source attribution pipeline that classifies sources using confidence-based flagging and LLM-as-judge check. Ambiguous cases are routed to business development through SeaTalk alerts, saving 5–10 manhours a month and raising data completeness for missing traffic sources by over 90%.
  • Extended the traffic dashboards with high-level user journey funnels for OKR reporting and tracking of internal and external marketing. Built product-level journey funnels that show stage-by-stage drop-off and separate good from bad traffic for segmented targeting.

Singapore Life (Singlife)

Senior Data Analyst (Analytics Engineering)

Apr 2026 – May 2026

  • Led delivery of a unified Actuarial ETL pipeline, replacing 6–9 months of annual manual effort with automated monthly runs, accelerating assumption validation across products and channels.
  • Coordinated cross-functional teams to architect an authoritative Single View of Advisor dataset — an end-to-end data pipeline from ingestion to standardised metrics — serving as the company-wide source of truth for advisor performance analytics.
  • Spearheaded the delivery of a simulation tool that models campaign costs, predicts advisor sales uplift under incentive structures, and quantifies ROI, enabling the business to dynamically simulate and optimise cash incentive campaigns.
  • Delivered an automated end-to-end data reporting pipeline that decomposes sales figures into actionable metrics, enabling downstream LLM-generated insights reporting via GenAI.

Data Analyst

May 2024 – Apr 2026

  • Built and deployed automated ETL pipelines (Python, SQL, Apache Airflow) on cloud infrastructure (AWS S3, Snowflake) to support reporting, data science, and GenAI use cases at scale, saving more than 200 manhours per month.
  • Conducted customer segmentation analyses on inforce policyholders to identify cross-sell gaps and quantify revenue potential, uncovering conversion opportunities and built a scalable framework for ongoing growth assessments.
  • Designed an account-level financial data model that unified fragmented P&L sources, shortened reporting cycles by 3 days, and enabled profitability insights that improved channel performance by up to 63.5%.

National Parks Board

Programmes Manager

Jun 2021 – May 2024

  • Applied statistical and population modelling techniques (oSCR, SECR, GLMM) in R to produce national-scale stray animal population estimates and derive predictors of stray animal density, guiding operational planning and policy.
  • Performed geospatial and statistical analysis using R and QGIS to identify spatial patterns in public feedback data and inform upstream mitigation strategies.
  • Managed cross-agency partnerships, led field operations, and delivered data-driven insights that shaped long-term animal management strategies.
  • Built the team’s first stray-dog TNRM capture dashboard in Qlik, tracking dog capture and sterilisation operations to support data-driven decisions and resource allocation.

Projects

Stock Price ETL Dashboard

A fully automated, end-to-end ETL pipeline emulating production data ingestion and analytics engineering using Apache Airflow. Daily stock prices are pulled from the Yahoo Finance API, stored in PostgreSQL, and transformed with SQL to compute moving averages for analysis. Results are surfaced through an interactive Streamlit dashboard, enabling end-users to explore trends, analyse performance, and consume insights via data report downloads.

PythonAirflowPostgreSQLDockerPandasStreamlitPlotlyYahoo Finance API

Wise New Currency Route Funnel Analysis

A product analytics project evaluating the feasibility and performance of Wise's new INR → USD currency route (hypothetical). Combined pre-launch demand estimation methods to assess market potential and profitability with post-launch funnel analysis across the first eight weeks to identify adoption patterns, conversion drivers, and friction points. Used Python and Pandas for data preparation, DuckDB for analytical querying, and Plotly for interactive visualisations, delivering actionable insights to inform product optimisation and strategic decision-making.

PythonDuckDBPandasPlotlyFunnel Analysis

Singapore Labour Market Analysis

Comprehensive analysis of Singapore’s labour market using MOM, OECD, and World Bank datasets. Built a full ETL workflow to retrieve and clean data, stored it in a MySQL database, and conducted multi-dimensional analysis across unemployment, retrenchment, turnover, wages, working hours, leave benefits, and flexible work arrangements. Visualisations and PCA were used to uncover insights on job security, sectoral opportunities, and work–life balance for job-seekers in a tight labour market.

PythonMySQLPandasScikit-learnSciPySeabornPlotlyETLPCA

Singapore Public Housing Market Analysis

Comprehensive analysis of Singapore’s public housing market using multiple HDB datasets (BTO prices, resale transactions, rental trends, housing grants). Performed data cleaning, feature engineering, and visualisation with NumPy and Matplotlib, generating insights on price patterns, estate affordability, financial support accessibility, and spatial housing trends across Singapore.

PythonNumPyMatplotlibGeoPandasPandasOneMap API

Campaign A/B Testing

A statistical analysis project testing whether an ad campaign lifts conversion against a PSA control group. Sized the sample with a power analysis, compared conversion rates with a two-proportion z-test, then used Mann-Whitney U, chi-square tests, and logistic regression to check how the number of ads seen and the day they were seen relate to conversion.

PythonstatsmodelsSciPyPandasSeabornA/B TestingPower AnalysisLogistic Regression

Applied Machine Learning Projects

Four machine learning projects from my Specialist Diploma in Data Science at Singapore Polytechnic. Predicted King County house prices with linear regression, random forest, and XGBoost, checking multicollinearity with VIF and mapping sales with GeoPandas. Forecast Apple, Amazon, and DBS stock prices with SARIMAX and Holt-Winters. Built a stacking classifier for Titanic survival, and compared K-Means, DBSCAN, Gaussian mixture, and agglomerative clustering on the UCI wine dataset.

PythonScikit-learnXGBooststatsmodelsGeoPandasTime Series ForecastingClustering

WhatCMS URL Enquiry App

A Python command-line tool that reads a list of URLs, queries the WhatCMS API for each site's CMS, programming language, database, CDN, and web server, and writes the results to CSV. Structured as a small package with a separate API client and enrichment layer, a dataclass response model, type hints, YAML config, a fixed delay between requests to stay within the API rate limit, and dependencies managed with uv.

PythonRequestsPandasREST APIuv

Education

Georgia Institute of Technology

Master of Science in Computer Science (Specialisation in Artificial Intelligence)

Aug 2025 – Present

  • Courses:

    • Machine Learning: Supervised and unsupervised models, plus randomised and gradient-based optimisation, practised as reproducible Kedro/MLflow pipelines rather than one-off notebooks.
    • Artificial Intelligence: Search, planning, probabilistic reasoning, and knowledge representation — the CS foundations underneath the applied ML work.
    • Software Development Process: Structured and agile delivery, QA, and lifecycle tooling; the engineering habits now used on data pipelines.
    • Agentic AI Essentials: Prompting, LLMs, RAG, and ReAct — the same building blocks as the GenAI products at Monee.
    • Introduction to C: Language, toolchain, and systems techniques; closing the gap from analyst tooling to how software actually runs.
    • Computer Networks: Signalling and framing, error control, medium access, routing, congestion control, end-to-end transport, and network APIs.
  • Built end-to-end, reproducible ML pipelines with Kedro, PyTorch, scikit-learn, MLflow, and Optuna — spanning supervised models, clustering/dimensionality reduction, and a from-scratch optimiser and randomised-optimisation study.

Singapore Polytechnic

Specialist Diploma in Data Science (Artificial Intelligence)

Apr 2024 – Mar 2025

  • Completed projects in applied machine learning, deep learning, time-series forecasting, generative AI, and human–AI interfaces.