Ahmad Fadli T.

Key Skills

Software

Don't disclose

CVAT

LabelImg

Label Studio

Roboflow

Top Subject Matter

Oil palm pollen object detection

image dataset labeling for agricultural plant health

Educational instruction (database)

Top Data Types

Image

Text

Top Task Types

Data Collection

Classification

Object Detection

Bounding Box

Segmentation

Question Answering

Polygon

Freelancer Overview

AI/ML Engineer (Intern) — Dataset preparation and labeling for object detection and LLM knowledge retrieval prototype (E. Brings 4+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Don't disclose, Internal, and Proprietary Tooling. Education includes Bachelor of Science, University (USU) (2022) and Professional Certificate, Bangkit Academy (2024). AI-training focus includes data types such as Image and Text and labeling workflows including Data Collection, Evaluation, and Rating.

Labeling Experience

AI Engineer - PT Riset Perkebunan Nusantara

ImageClassificationObject Detection

Developed core RAG and agent workflows for Chat OPA, an AI assistant for oil palm information and plant health analysis. Built a FastAPI backend covering knowledge retrieval, query processing, and response generation with computer vision integration. Integrated disease and pest detection models for image-based plant analysis while supporting production delivery across web and mobile platforms using Python and LLM tooling. • Implemented LangChain and LangGraph agent/RAG pipelines • Designed and deployed FastAPI services for LLM-backed retrieval and generation • Integrated computer vision models for conversational plant image analysis • Managed field image collection and supervised annotation quality with iterative training and evaluation

2025 - 2026

AI Engineer — RAG/agent workflows with supervised image dataset labeling and iterative model training (Chat OPA)

Don't discloseImageData Collection

Collected field images and supervised an annotation team to maintain labeling quality and consistency for plant health analysis. Developed RAG and agent workflows for a conversational assistant that integrates knowledge retrieval and computer vision in production. Performed iterative training and evaluation using dataset refinement to improve model performance for integrated disease and pest detection. • Collected 10,000+ field images for labeling • Supervised annotation quality to ensure consistent labels • Integrated disease/pest detection outputs into an AI assistant workflow • Conducted iterative dataset refinement and evaluation to improve performance

2025 - 2026

AI/ML Engineer Intern - Indonesian Oil Palm Research Institute

TextClassificationObject Detection

Developed an object detection system for classifying healthy vs dead oil palm pollen and computing pollen quality ratios. Prepared and labeled datasets, supervised annotation quality, trained the model, and created a Streamlit interface for visualization of results. Built ElAiS, an LLM-powered assistant using Hybrid RAG with LangChain and Neo4j to support company knowledge retrieval for a research workflow. • Built object detection pipelines and Streamlit-based result dashboards • Implemented Hybrid RAG using LangChain with Neo4j knowledge storage • Prepared labeled datasets and supervised labeling quality checks • Delivered an LLM assistant prototype that later informed subsequent product work

2024 - 2024

AI/ML Engineer (Intern) — Dataset preparation and labeling for object detection and LLM knowledge retrieval prototype (ElAiS)

Don't discloseImageData Collection

Prepared and labeled datasets for an oil palm object detection system to classify healthy versus dead pollen and derive pollen quality ratios. Supervised annotation quality and trained the detection model using the labeled dataset. Built an interface to visualize model outputs and assist downstream evaluation. • Prepared and labeled image datasets for detection • Supervised annotation team/quality to maintain consistency • Trained the object detection model on labeled data • Visualized results via Streamlit for review and validation

2024 - 2024

Laboratory Assistant — Course practicum teaching

Image

Supported practical teaching activities for a course focused on advanced database management rather than AI training, data labeling, or annotation tasks. There was no explicit AI labeling, annotation, or dataset generation responsibility described in the role. Therefore, this entry is included only as indirect training support with no identifiable labeling artifacts. • Delivered course practicum guidance • Managed lab instruction for students • Covered database concepts in a training context • No explicit data labeling/annotation performed

2023 - 2024

Education

Work History