Machine Learning Engineer - Remote - Across Southeast Asia

NPAworldwide; Makati; On-site; PHP 1,800,000 - 3,000,000; Full time

Job summary

NPAworldwide in Makati, Manila seeks a Senior Data Scientist to own end‑to‑end statistical and ML analysis for its decision‑intelligence platform. You frame problems as modelling tasks, engineer features, and drive production monitoring and retraining.

You will work across discovery, solution workshops, and stakeholder engagements, defending model assumptions and outputs to both technical and executive audiences, while advancing LLM/RAG and MLOps capabilities within the platform.

Qualifications

• 6+ years building and shipping data science / ML in Python with production ownership.
• Strong statistics, Bayesian inference, time-series, and causal reasoning.
• Proven model selection, feature engineering, and evaluation skills.
• Production-grade software engineering fundamentals and API design.
• Experience with MLOps, monitoring, and retraining in production.
• Excellent client-facing communication and workshop facilitation.

Responsibilities

• Own, manage, enhance, and support the decision‑intelligence platform.
• Direct model selection, feature engineering, and evaluation across engagements.
• Frame business problems as ML tasks with metrics, data needs, and validation.
• Design, deploy, monitor ML/AI services from data to API and dashboards.
• Generalise solutions across domains and integrate LLM/RAG as needed.
• Implement MLOps practices including versioning and drift detection.
• Act as client-facing technical expert during pre-sales, discovery, and delivery.
• Deliver presentations, workshops, and findings to technical and executive audiences.
• Maintain standards for model quality, reproducibility, and responsible AI.

Skills

Data science

Python

Statistics

Bayesian inference

Time-series forecasting

Causal inference

ML deployment

MLOps

LLMs / RAG

Vector databases

SQL

FastAPI

Tools

PostgreSQL

ClickHouse

Pinecone

QDrant

FAISS

Azure AI Search

Job description

Job Description

Job Title: Senior Data Scientist

Job Family: Data Scientist

Level: Level 2 Senior

Reports To: Chief Operating Officer (COO) / Head of AI Product Development

Department: AI Product Development

Role Summary

The Senior Data Scientist owns the end-to-end statistical and machine learning analysis behind Company's decision‑intelligence platform from framing a business problem as a modelling question, through feature engineering and model evaluation, to production monitoring and retraining. The role combines strong applied statistics and causal reasoning with hands‑on delivery experience, and acts as the client-facing technical expert across discovery, solution workshops, and stakeholder engagements explaining and defending model assumptions and outputs to technical and non‑technical audiences alike. Industry experience in pharmaceuticals or life sciences is highly regarded.

Platform

A decision‑intelligence platform that tells operators not just what happened, but why it happened, what to do about it, and what will happen if they act across seven industries, from a single shared intelligence core.

Most analytics tools stop at the dashboard: they show you the numbers and leave the interpretation to you. Our platform closes that gap. It reasons over the operation the way an expert analyst would finding the real drivers behind a result, recommending the action most likely to improve it, and simulating the consequences before committing.

The Senior Data Scientist contributes to this platform as the statistical and modelling foundation of Company's AI/ML capability.

Key Responsibilities

• Own, manage, enhance, and support Companys decision‑intelligence platform as the foundation of AI/ML delivery.
• Own model selection, feature engineering, and evaluation strategy across client engagements.
• Frame business problems as ML problems, defining success metrics, data requirements, and validation approaches.
• Design, build, deploy, monitor, and support ML and AI services in production from data ingestion through to modelling, API, and dashboard validation.
• Generalise solutions across problem domains rather than building one-off implementations.
• Integrate LLM and RAG capabilities where they add value, applying prompt engineering and fine‑tuning as required.
• Implement MLOps practices for model versioning, monitoring, retraining, and performance management.
• Act as the client-facing technical expert during pre‑sales, discovery, and delivery phases.
• Deliver client presentations, solution workshops, and stakeholder engagements, presenting findings, prototypes, and recommendations to technical and executive audiences.
• Establish and uphold standards for model quality, reproducibility, responsible AI, and engineering rigour (cutover docs, adversarial testing, safety checks).
• Contribute to Companys data science and AI/ML capability.
• Produce clear technical documentation, solution designs, and best‑practice guidance.

Qualifications

• 6+ years building and shipping data science / applied ML work in Python, with direct ownership of analysis through to production monitoring not notebooks-to-handoff.
• Strong applied probability and statistics: Bayesian inference (conjugate updates, posterior fitting, hierarchical models), time‑series forecasting, and a working grasp of causal inference (reasoning about confounding, identification, and bias in fitted effects).
• Proven expertise in model selection, feature engineering, feature selection, model evaluation, and validation strategies, including crossvalidation, bias/variance management, train‑test design, and drift detection.
• Production‑grade software engineering fundamentals: typed Python, pytest, SQL (PostgreSQL), and comfort with at least one columnar/OLAP store (ClickHouse a plus).
• Ownership of ongoing model performance monitoring, drift detection, and retraining cadence for models in production.
• Demonstrated ability to generalise modelling approaches across multiple problem domains rather than special‑casing each engagement.
• Excellent client‑facing communication, presentation, and workshop facilitation skills; able to translate business problems into ML/DS problems and explain results to non‑technical and executive audiences.
• Ability to explain and defend model assumptions, methodology, and outputs to sceptical or non‑technical stakeholders, including handling pushback on AI/ML‑driven recommendations.
• Sound understanding of data governance, security, and compliance considerations relevant to client engagements.
• Rigour as a default: writes the cutover doc, adds the adversarial test, and does not bypass validation to make an obstacle go away Strongly Preferred
• Practical experience with LLMs, embedding models, and retrieval‑augmented generation (RAG), including prompt engineering and fine‑tuning.
• Strong background designing and optimising vector databases (Pinecone, QDrant, FAISS, Azure AI Search) for semantic search and RAG systems.
• CI/CD gating, schema migrations, and API design (FastAPI or equivalent) i.e. service ownership beyond the model itself.
• Incident response experience for production ML services.
• Online and sequential decision‑making: Thompson sampling, multi‑armed bandits, Bayesian RL, or contextual bandits in production.
• Causal‑discovery and attribution tooling: DoWhy, Tigramite/PCMCI+, structural‑VAR, IV methods.
• Econometric and Bayesian fitting libraries: PyMC, statsmodels, pymc‑marketing.
• React/TypeScript literacy enough to own the backend‑to‑frontend contract and validate UI end‑to‑end
• Pharmaceutical or life sciences domain experience (clinical data, real‑world evidence, drug discovery, commercial analytics, or regulated GxP/HIPAA environments).

Why Is This a Great Opportunity

• Currently 30 to 40 people, intentionally kept lean for several years
  • Deliberately flat structure: no rigid hierarchy, project‑based team assembly, entrepreneurial culture
  • Senior leadership team described as exceptionally experienced; competes with top‑tier consultancies
  • Current gap: thin middle layer between senior leadership and strong technical/junior staff
  • The two new hires are intended to be the glue that closes this gap
• Growth trajectory: targeting 100 to 200 people over the next two years
  • Significant US headcount build planned alongside Southeast Asia hiring
  • Government‑related contracts require US‑based staff; non‑US candidates excluded from those workstreams
  • Organic growth model: no external funding, deliberate and measured pace