AI Engineer with less than a year in AI pipeline development and data engineering.
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Motivated and results-driven AI Engineer (Intern/Part-time) with 11 months of experience in developing end-to-end deep learning pipelines, data engineering, and real-time AI systems. Proficient in PyTorch, TensorFlow, and various data processing tools. Proven ability to implement advanced retrieval strategies for RAG systems and optimize speech translation pipelines, demonstrating strong problem-solving and technical skills in artificial intelligence and machine learning.
Danang University of Science and Technology
Bachelor of Information Technology · Information Technology
August 1, 2023 – Present
VJ Technologies
AI Engineer (Intern → Part-time)
November 1, 2025 – June 1, 2026
India
MakeAI
Software Engineer (Intern)
July 1, 2025 – September 1, 2025
India
Institutional Knowledge RAG System (DUT-RAG)
March 1, 2026 – May 1, 2026
Designed and deployed a full-stack RAG system processing 1,000+ multi-domain institutional documents (legal files & web news) utilizing automated web crawlers and OCR pipelines. Implemented a multi-stage retrieval architecture combining Sparse (BM25) and Dense Embeddings, integrated with Reciprocal Rank Fusion (RRF) and cross-encoder Rerankers (Cohere, local BGE), achieving an exceptional 95-100% Recall@5. Deployed a high-performance system serving 16,400+ data chunks, delivering 95-100% retrieval recall, <50ms search response time, and <1.5s LLM generation latency.
SpeechBridge - End-to-end Speech Translation Pipeline
February 1, 2026 – March 1, 2026
Designed and developed an end-to-end English-to-Vietnamese speech translation system integrating Whisper (ASR), M2M100 (NMT), and SpeechT5 (TTS) for seamless cross-lingual communication. Fine-tuned Whisper-small using 4-bit QLORA on cloud GPUs, enabling parameter-efficient training while maintaining high transcription quality. Built a scalable parallel audio preprocessing pipeline with VTT parsing and ±150ms boundary padding to improve speech-text alignment and dataset quality. Achieved a 10.5% relative reduction in both Word Error Rate (WER) and Character Error Rate (CER) compared to the baseline Whisper-small model.
Vietnamese Abstractive Summarization
September 1, 2025 – December 1, 2025
Developed an end-to-end Vietnamese abstractive summarization system using a custom Transformer architecture in PyTorch for automated news article summarization. Built a scalable data pipeline to collect and process 438K+ samples from the VietNews-Abs-Sum dataset, including custom Vietnamese BPE tokenization and Unicode normalization. Implemented encoder-decoder Transformer with multi-head attention and positional encoding, evaluated using ROUGE, F1-score, and accuracy metrics.
Cultural Fit Analysis
The candidate's academic projects (DUT-RAG, SpeechBridge, Vietnamese Abstractive Summarization) demonstrate a strong interest and practical application in AI/ML, aligning well with an AI Engineer role. The internship experiences at VJ Technologies and MakeAI further solidify this alignment, showcasing real-world application of AI principles and system development. The diversity of projects, from RAG systems to speech translation and geospatial AI, indicates a broad skill set and adaptability, which are positive indicators for cultural fit in a dynamic AI environment. The focus on performance optimization and scalable solutions suggests a results-oriented mindset. The candidate is currently pursuing a Bachelor's degree, which might indicate a preference for roles that support continued learning and growth.
Soft Skills & Operational Fit
The candidate's project descriptions highlight strong problem-solving skills, particularly in optimizing complex AI systems for performance and recall. The individual project on Vietnamese Abstractive Summarization demonstrates initiative and independent work capability. The collaborative nature of the RAG and Speech Translation projects suggests good teamwork potential. The detailed descriptions of technical challenges and solutions indicate a methodical and analytical approach to engineering. However, without direct interview data, assessing communication clarity, stress handling, and team collaboration in a live operational setting is limited.