Senior AI Engineer - Digital Onboarding for Financial Institutions

At Boost, we build the AI infrastructure powering digital onboarding for banks and microfinance institutions across Southeast Asia.

We enable Financial Institutions to onboard clients digitally for loans and savings in 5–10 minutes — through chat, without app downloads.

Traditionally, banking meant physical branches and long forms. Boost turns that on its head — with chat-based lending, document AI, and agentic workflows that make financial services 100x faster and radically more accessible.

We integrate with a new partner bank in 2–3 weeks, helping them expand their reach instantly.

Backed by top-tier institutional and angel investors, Boost is building the digital rails for inclusive finance in emerging markets.

Field: AI Engineering

Position: AI Engineer

Location: Work From Anywhere (teams across PH, EU, and APAC)

Assignment Category: Full-time

Reporting Line: Head of AI Engineering

• Build, train, and deploy deep learning models for OCR, document fraud detection, and anomaly classification.
• Optimize multimodal pipelines combining visual (layout, handwriting, stamps) and textual (LLM) inputs.
• Fine-tune large language models and vision models (CLIP, Donut, LayoutLM, Phi, etc.) for financial document understanding.
• Create evaluation frameworks to measure model precision, recall, and fraud detection accuracy across document types.

• Design pipelines for large-scale data ingestion, annotation, and feature extraction across multiple document formats (PDF, image, text).
• Implement data quality checks, balancing datasets for bias, completeness, and fraud pattern diversity.
• Work closely with ML Ops and backend teams to version datasets, maintain lineage, and automate retraining triggers.

• Deploy models as scalable APIs on GCP using Vertex AI, Cloud Run, or GKE.
• Build real-time inference systems with caching and rate limiting for high-throughput banking environments.
• Integrate model outputs with backend microservices and downstream analytics systems.

• Explore state-of-the-art approaches in document AI, fraud detection, and multimodal learning.
• Prototype and benchmark new architectures, from transformer-based OCR models to graph-based document linking.
• Collaborate with the product team to translate business requirements into measurable AI outcomes.

• Work cross-functionally with AI Ops, backend, and product engineers to bring research into production.
• Contribute to internal tools that automate data labeling, model monitoring, and explainability.
• Communicate complex technical findings clearly to both engineers and non-technical stakeholders.

Qualifications

• 5+ years of experience developing and deploying ML models in production.
• Strong background in Python, PyTorch or TensorFlow, and experience with cloud AI platforms (Vertex AI, AWS Sagemaker, or similar).
• Familiarity with API Development frameworks like FastAPI
• Experience in computer vision (OCR, layout analysis, document parsing) and NLP (entity extraction, summarization).
• Familiarity with model optimization techniques (quantization, distillation, batching).
• Understanding of CI/CD for ML (GitHub Actions, Cloud Build, containerization).
• Comfort with GCP tools such as BigQuery, Pub/Sub, and Dataflow.
• Strong analytical and debugging skills — able to connect model metrics to business KPIs.
• Bonus: Experience with fraud detection, anomaly detection, or financial document data.

What we Offer

• Full work from home 
• Competitive salary 
• PH dayshift (9am to 6pm)

• Fully AI-powered internal workflows that actually work