Vetted

Anita W.

Senior AI Engineer

Experience

5+ Years

Availability

Full-time

Location

🇮🇩 Jakarta, Indonesia

English

Advanced

Python PyTorch LangChain Hugging Face

Hire Anita

Verified by Second Talent Jan 2024

🇮🇩 Hire developers in Indonesia →

About

Anita is an AI engineer who builds production-grade machine learning systems and LLM-powered applications. She has shipped ML features used by millions of users across e-commerce, healthtech, and fintech platforms. Her expertise spans the full ML lifecycle from data pipeline design through model training, fine-tuning, and deployment. She is deeply experienced with modern LLM frameworks including LangChain and Hugging Face Transformers, and has deployed RAG systems, AI agents, and custom fine-tuned models in production environments.

Languages

English

Advanced

Bahasa Indonesia

Native

Skills

AI/ML

PyTorch TensorFlow Hugging Face scikit-learn ONNX

LLM

LangChain OpenAI API Claude API RAG Fine-tuning

Backend

Python FastAPI Flask Celery

Data

PostgreSQL Pinecone Weaviate Redis Apache Spark

Cloud

AWS SageMaker GCP Vertex AI Docker Kubernetes

Experience

Senior AI Engineer

2023 - Present

Xendit

Leading AI feature development including a RAG-powered knowledge assistant and automated content moderation system. Deployed LLM agents that handle 30K+ queries daily with 94% accuracy.

LLM agents handle 30K+ queries/day with 94% accuracy, reducing support tickets by 45%
RAG knowledge assistant indexes 200K+ documents with sub-2s retrieval latency
Content moderation system processes 50K+ user posts/day, flagging policy violations with 96% precision

Machine Learning Engineer

2021 - 2023

Tokopedia

Built recommendation systems and search ranking models serving 12M+ monthly users. Implemented real-time fraud detection pipeline using PyTorch and Apache Kafka.

Recommendation engine increased average order value by 18% across 12M monthly users
Fraud detection pipeline caught $2.3M in fraudulent transactions in first 6 months
Search ranking model improved click-through rate by 32%

Data Scientist

2020 - 2021

Halodoc

Developed predictive models for patient risk scoring and clinical trial matching. Reduced false positive rates by 40% through ensemble model optimization.