Nikola Rahman, Developer in Belgrade, Serbia
Nikola is available for hire
Hire Nikola

Nikola Rahman

Verified Expert  in Engineering

Machine Learning Developer

Location
Belgrade, Serbia
Toptal Member Since
August 26, 2021

Nikola is a highly skilled ML expert with over eight years of experience. He has extensive knowledge in various business and data domains. He has worked on a wide range of projects, from training cutting-edge models for startups to delivering solutions for industry leaders. Nikola has a proven track record of building successful systems from scratch. With his expertise in building machine learning systems from the ground up, he is well-equipped to tackle any challenge that comes his way.

Portfolio

Stop the Traffik
Amazon SageMaker, Natural Language Processing (NLP), OpenAI GPT-3 API...
Site Matrix, LLC
Machine Learning, Artificial Intelligence (AI), Big Data, Data Analysis...
Signaloid Limited
Machine Learning, Data Science, C, Signal Processing, Python, Pandas...

Experience

Availability

Part-time

Preferred Environment

Jupyter Notebook, Linux, Jira, Slack, Visual Studio Code (VS Code)

The most amazing...

...system I've built estimates fetal heart rate from abdominal electrodes. It's accurate, reliable, and provides peace of mind for expectant parents.

Work Experience

Data Scientist

2023 - 2023
Stop the Traffik
  • Developed a news article classification model that is used to find articles related to modern human slavery and human trafficking (MHSHT). The organization uses the model to monitor the MHSHT realm and derive reports from it continuously.
  • Achieved human-level accuracy of the model. The model has 99% recall and 75% precision.
  • Deployed the model to a Kubernetes cluster on IBM Cloud for daily processing.
Technologies: Amazon SageMaker, Natural Language Processing (NLP), OpenAI GPT-3 API, OpenAI GPT-4 API, PyTorch, Data Science, Generative AI

ML Developer

2023 - 2023
Site Matrix, LLC
  • Built a revenue forecast model for a specific public company using a scikit-learn library that outperforms analyst estimates by an order of magnitude.
  • Deployed the model training and inference pipeline for continuous improvement and simple use.
  • Performed data analysis on hundreds of terabytes of data using AWS Athena, SQL, and scikit-learn to come up with model features.
Technologies: Machine Learning, Artificial Intelligence (AI), Big Data, Data Analysis, Data Science, Predictive Analytics, Probability Theory

Data Scientist

2022 - 2023
Signaloid Limited
  • Ported a Bayesian neural network (BNN) for precipitation prediction from TensorFlow GPU for a specialized hardware platform. Implemented neural network layers in C from scratch.
  • Developed standard digital signal processor (DSP) and ML functions in C, like Radix-2 FFT, Levenberg-Marquardt algorithm, non-linear least squares optimization, IIR filter design procedure, and filtering, among others.
  • Ported a signal-processing pipeline for underwater cable displacement estimation from MATLAB to C.
Technologies: Machine Learning, Data Science, C, Signal Processing, Python, Pandas, Convolutional Neural Networks (CNN), Predictive Analytics, Probability Theory

Data Scientist and NLP Engineer

2021 - 2022
Law of the Jungle Pty Limited
  • Developed a POC for classifying claims in marketing campaigns to decrease the time spent on marketing compliance activities.
  • Used semantic search and other NLP techniques to achieve a 50% improvement in accuracy over the existing system.
  • Provided the client with comprehensive guidance and strategic recommendations, enabling them to proactively align their AI strategy and optimize their data collection activities in preparation for a successful business transformation through AI.
Technologies: Python, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT), OCR, Data Science

Lead Machine Learning Engineer | Project Manager

2018 - 2021
HTEC Group
  • Successfully founded and led a machine learning team at my company, growing the team from 0 to 20 members.
  • Developed a certified medical-grade ECG analysis software that has been adopted in several clinics in the healthcare industry.
  • Completed over 20 projects for both startups and established blue-chip companies, working in R&D to build POCs and advance the state-of-the-art in the field.
Technologies: Python, Deep Learning, Machine Learning, Data Science, MATLAB, Signal Processing, Optimization, Technology Consulting, Advisory, Startup Consulting, Artificial Intelligence (AI), OpenCV, Computer Vision Algorithms, Software Development, Spreadsheets, Predictive Modeling, NumPy, Pandas, Matplotlib, Plotly, Amazon Web Services (AWS), Deep Neural Networks, Seaborn, AI Programming, Natural Language Understanding (NLU), GitHub, Data Visualization, Cloud, JupyterLab, Convolutional Neural Networks (CNN), Probability Theory, Generative AI

ML Engineer

2017 - 2018
Freelancer
  • Launched the POC Android app successfully, which uses a trained image classification CNN model to perform real-time product classification in the supermarket with high accuracy.
  • Collaborated with a team of experts to collect and label a comprehensive dataset of 10,000 images, which was used to train the model.
  • Optimized the model's performance by using techniques such as transfer learning and data augmentation, achieving real-time performance on the Android app.
  • Conducted thorough testing and debugging to ensure the app's stability and reliability in real-world scenarios.
Technologies: TensorFlow, Deep Learning, Keras, Google Cloud, Image Processing, Computer Vision, Artificial Intelligence (AI), Machine Learning, Computer Vision Algorithms, Software Development, Statistics, Spreadsheets, NumPy, Pandas, Matplotlib, Plotly, Amazon Web Services (AWS), Deep Neural Networks, Seaborn, AI Programming, Natural Language Understanding (NLU), GitHub, Data Visualization, Cloud, JupyterLab, Convolutional Neural Networks (CNN), Data Augmentation

Senior ML Engineer

2016 - 2018
HTEC Group
  • Pioneered the use of novel deep learning architecture for event-based vision cameras, reducing FLOPS by a staggering 80%.
  • Spearheaded the development of five cutting-edge algorithms for a client's deep learning library in the highly efficient Halide language.
  • Engineered convolutional neural networks for lightning-fast inferences on mobile devices, achieving a jaw-dropping 40x speed boost with advanced techniques like quantization, pruning, and expert architecture design.
Technologies: Python, SQL, Artificial Intelligence (AI), Deep Learning, Machine Learning, TensorFlow, Keras, PyTorch, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT), Image Processing, Computer Vision, Software Development, Statistics, Linear Regression, Spreadsheets, Predictive Modeling, NumPy, Pandas, Matplotlib, Amazon Web Services (AWS), Deep Neural Networks, Seaborn, AI Programming, Natural Language Understanding (NLU), GitHub, Data Visualization, Cloud, JupyterLab, Convolutional Neural Networks (CNN)

ML Engineer

2014 - 2016
HTEC Group
  • Developed medical-grade algorithms for arrhythmia classification, beat classification, and ECG morphology analysis that help doctors in early detection of heart disease.
  • Built highly-optimized ECG signal processing pipelines in C.
  • Wrote tests for ECG algorithm performance according to ISO 60601-2-47 and ISO 60601-2-25 standards.
Technologies: MATLAB, C, Python, Linux, Theano, TensorFlow, Artificial Intelligence (AI), Machine Learning, Software Development, Spreadsheets, Predictive Modeling, NumPy, Pandas, Matplotlib, Amazon Web Services (AWS), Deep Neural Networks, Seaborn, AI Programming, Natural Language Understanding (NLU), GitHub, Data Visualization, Cloud, JupyterLab, Convolutional Neural Networks (CNN)

Advancing ECG Analysis with Signal Processing and Machine Learning

Revolutionizing ECG Analysis with Cutting-Edge Signal Processing and Machine Learning

I developed signal processing and ML pipelines for ECG processing. I worked with a cardiology expert to develop medical-grade algorithms for several types of ECG analyses and arrhythmia detection models. I was one of the first to apply deep learning to ECG signals. The algorithms comply with essential performance requirements found in EN 60601-2-25 and EN 60601-2-47 standards and are used daily in several clinics for the early detection of heart disease.

High-performance Deep Learning Library for a Low-power DSP

I designed and implemented cutting-edge machine learning algorithms and neural network layers using the novel Halide programming language. These algorithms and layers were specifically optimized for high processing speed and low power consumption, making them ideal for use in real-time applications. My contributions to the client's deep learning library included three advanced machine learning algorithms, four state-of-the-art convolutional neural network layers, and an algorithm for neural network quantization. These additions significantly enhanced the library's capabilities and demonstrated my expertise in the field of artificial intelligence and deep learning.

Real-Time Image Segmentation on a Snapdragon 820 Platform with Improved CNN Model

I collaborated with a smart home appliances company to improve their CNN model for real-time image segmentation on a Snapdragon 820 platform. Through my efforts, the model achieved a 50x speed-up while maintaining 99% accuracy. This improvement allows the company to use the model for more efficient and effective image analysis in their smart home products.

Unleashing the Power of Event-Based Cameras with Custom Spiking Neural Networks

The project involved collaborating with a chip design company to implement their custom spiking neural network architecture for processing event-based camera data. I developed custom TensorFlow layers to support the implementation and benchmarked the neural net on known event-based datasets. This enabled the client's research team to experiment with novel neural net architectures optimized for their chip architecture, advancing their research in this area.

Deep Learning for Automatic Podcast Topic Breakdown

I developed deep learning models for an automatic topic breakdown in podcast episodes. Working with a team of 20 engineers and data collectors, we labeled and trained our models on 10,000 episodes and achieved human-level performance. Our algorithm enables users to easily navigate topics within podcast episodes and listen to segments of interest.

Non-Invasive Fetal ECG Recording with Dry Electrodes

I developed a medical-grade system for non-invasive recording of fetal ECG from the maternal abdomen. I designed experiments to determine optimal electrode placement for maximum signal capture and quality and created a signal-processing pipeline for estimating maternal and fetal heart rates and uterine activity. This is the first system of its kind to use dry electrodes, making it easy for moms to record fetal ECG without any skin preparation. The system offers higher signal capture and accuracy than traditional methods like CTG and is comparable to similar systems that use adhesive electrodes.

XAI-Enhanced Diabetic Retinopathy Classification with Explainable Insights

I have developed an explainable AI (XAI) method for a deep-learning computer vision model that classifies diabetic retinopathy. The method provides stakeholders with insights into the model's predictions, allowing them to better understand how and why the model makes its decisions.

Accelerating Deep Learning Classification with Cloud-Based GPU Clusters

I effectively scaled the training of a deep learning classification model from a single on-premise node to a cloud-based 100+ GPU node cluster. By utilizing advanced techniques such as DistributedDataParallel and automatic mixed precision training, I significantly reduced the model's training time from weeks to less than 24 hours. This enabled faster experimentation and iteration, improving model performance and accuracy.

ECG Classification Improvements with Deep Learning

I led the client's small team of machine learning specialists and helped improve their deep learning models for ECG classification. Through my guidance and implementation of more advanced techniques, we were able to reduce the error rate by over 40%. In addition, I provided expertise on FDA compliance and improved their model validation pipelines. Overall, my leadership and technical knowledge allowed for significant progress in the development of their machine-learning models.
2010 - 2014

Bachelor's Degree in Electrical Engineering

University of Belgrade - Belgrade, Serbia

Libraries/APIs

TensorFlow, PyTorch, Keras, SciPy, Scikit-learn, NumPy, Pandas, Matplotlib, OpenCV, Theano

Tools

MATLAB, Slack, Seaborn, Plotly, MATLAB Neural Network Toolbox, GitHub, ChatGPT, Snapdragon Neural Processing Engine (SNPE), Git, Jira, Spreadsheets, Jenkins, Apache Airflow, Amazon SageMaker

Languages

Python, Bash, SQL, C

Paradigms

Data Science

Platforms

Jupyter Notebook, Linux, Visual Studio Code (VS Code), Amazon Web Services (AWS), AWS Lambda, Docker, Amazon EC2

Storage

Amazon S3 (AWS S3), Google Cloud

Other

Signal Processing, Deep Learning, Machine Learning, Optimization, Digital Signal Processing, Event-based Vision, Image Processing, Audio Processing, Natural Language Processing (NLP), Computer Vision, Text Classification, Text Categorization, Artificial Intelligence (AI), Text Analytics, Neural Networks, Deep Neural Networks, Computer Vision Algorithms, Linear Regression, Predictive Modeling, AI Programming, Natural Language Understanding (NLU), Hugging Face, Data Visualization, JupyterLab, Generative Pre-trained Transformers (GPT), Convolutional Neural Networks (CNN), OpenAI GPT-3 API, OpenAI GPT-4 API, Predictive Analytics, Probability Theory, Explainable Artificial Intelligence (XAI), OpenAI, Generative AI, Software Development, Information Retrieval, Probabilistic Information Retrieval, Statistics, Technology Consulting, Advisory, Startup Consulting, Cloud, Halide, Speech to Text, Distributed Systems, Video Processing, Data Mining, Big Data, Data Analysis, OCR, Data Augmentation

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring