Speech Emotion Recognition using Transfer Learning Approach and Real-Time Evaluation in English and Bengali Language
Shanjidul Islam Sadhin. (2023). "Speech Emotion Recognition..." Pre-Print.
CV
Education
- Bachelor of Science in Computer Science & Engineering
- American International University-Bangladesh (AIUB)
- January 2020 - December 2023
- Distinction: Dean’s List Award (August 2023)
Work Experience
- Applied AI Engineer (Remote)
Askturing.ai June 2025 - Present - Reduced LLM token costs by 40% by designing an intelligent context management and chunking strategy.
- Developed a vision retrieval system for optimized, cost-efficient information extraction.
- Transitioned from text-based LLMs to Vision-Language Models (VLMs).
- Architected and refined a RAG evaluation framework.
- Implemented an ephemeral file management feature in the Askturing Chat application.
- Developed multi-step hybrid reasoning workflows.
- Implemented AI Faithfulness & Fact-Validation pipelines.
- Implemented an event-driven action trigger system.
- AI Engineer (On-site)
TechnyX.ai December 2024 - May 2025 - Designed and architected an AI system for image generation and virtual try-on.
- Configured and optimized NVIDIA Triton Inference Server.
- Led development of advanced computer vision and multimodal AI solutions.
- Architected and deployed vision-language models for in-house API development.
- Freelancer (Remote)
UpWork October 2024 - Developed a state-of-the-art super-resolution model for image up-scaling.
- Modified the architecture of the “Dual Attention Transformer”.
- Designed a robust architecture that utilizes channel-wise concatenation.
- Junior AI Developer (Remote)
CRTVAI April 2024 - September 2024 - Designed and implemented system architecture and API endpoints for automated quality control.
- Fine-tuned speech-to-text models, deploying server-less models.
- Improved the Voice Activity Detection pipeline.
- Achieved a 25% reduction in GPT API calling costs.
- Research Assistant (On-site)
Computer Vision and Bio-Robotics Sep 2023 - Jan 2024 - Contributed to the creation of a research dataset containing 12,000+ labeled images.
- Reduced research time by 20% by implementing efficient data analysis.
Skills
- Languages: Python, C++, C#, JavaScript, PHP, SQL
- Libraries: Pytorch, TensorFlow, Keras, NumPy, Pandas, Scikit-Learn
- Frameworks: FastAPI, LangChain, Flask, Django
- Tools: GIT, DVC, MLflow, DagsHub, MySQL, PostgreSQL, ChromaDB, FAISS
- Platforms: Linux, Windows, Docker
- Cloud-Platforms: AWS, vLLM
- Soft Skills: Leadership, Writing, Public Speaking, Time Management
Publications
Talks
Teaching
Service and Leadership
- Problem Solving: Solved 800+ problems in different Online Judges.
- 12th National Undergraduate Mathematics Olympiad 2021: Participated (Dhaka North Region).
- Hacktoberfest-2023: Contributed to Refactoring and Programming Solutions.