Suleman Khan
Verified Expert in Engineering
Artificial Intelligence Engineer and Developer
Suleman has over five years of expertise in data engineering, machine learning, cloud computing, and back-end software development. He has a master's degree in data science and has published three research articles on interpretable machine learning. Suleman currently works on cutting-edge technologies, including Python, Docker, Kubernetes, AWS, Redis, Ray.io, FastAPI, GraphQL, Boto3, RabbitMQ, Celery, SQL, PostgreSQL, Prefect, TensorFlow and most of the python frameworks.
Portfolio
Experience
Availability
Preferred Environment
Python, Back-end, Artificial Intelligence (AI), Machine Learning, TensorFlow, Docker, PostgreSQL, Amazon Web Services (AWS), Cloud Computing, Kubernetes
The most amazing...
...project I've developed is a no-code data infrastructure platform to store, manage, process, and analyze the wind and solar plants data.
Work Experience
Applied Data Scientist | Machine Learning Engineer
ICS Collections
- Designed and developed a multi-tenant SaaS platform back end utilized by numerous medical debt collection clients.
- Integrated back-end APIs with ChatGPT to create intelligent reporting agents, optimizing data-driven decision-making.
- Integrated voice-based machine learning models and NLP semantic segmentation models to facilitate better decision-making processes.
- Developed an automated process for loading data from over 50 clients, streamlining data management workflows.
- Implemented a microservices architecture using FastAPI with Postgres, contributing to revenue growth exceeding $200,000.
- Designed and implemented a PostgreSQL database utilized by multiple agencies, scaling to over 50 million rows.
Data Scientist | Machine Learning Engineer
Altosphere
- Developed the back end of a visual data infrastructure platform to store, manage, and process data.
- Created unsupervised machine learning models and rule-based anomaly detectors for time series data.
- Built an information extraction and verification pipeline from PDF field reports using Amazon Textract and Boto 3.
- Utilized Ray.io to streamline the processing of PDF field reports, integrating with Amazon Textract and Boto 3 for efficient data extraction and verification.
- Implemented a dynamic data pipeline orchestration module using Prefect.
- Developed asynchronous task execution using Celery, RabbitMQ, and Redis. Monitored tasks with Flower.
- Integrated multiple microservices using GraphQL and gRPC. Containerized the FastAPI back end using Docker and deployed it on AWS.
Data Scientist / Machine Learning Engineer
Freelance Clients
- Developed an application to detect the number of times the advertisement is played during the radio transmission using TensorFlow.
- Detected the anomalies in the time-series dataset using CNN, RNN, and other anomaly detection algorithms.
- Created IVR using bidirectional Twilio streaming, Google speech-to-text, and text-to-speech services for a call center.
- Built an Android application using TensorFlow Lite to detect a person with COVID-19 in the voice recordings.
- Implemented a machine learning algorithm to detect a person's fall in a room CCTV recording.
- Consolidated data from multiple sources in Google BigQuery.
Software Engineer
Luminogics
- Built a website for an automobile company using HTML, CSS, JavaScript, React, and Node.js.
- Developed simple web games using the Phaser library.
- Created the back end of a website using Python, SQL, PostgreSQL, and Docker.
Software Engineer | Internship
Nextbridge
- Built a social application using Android networking libraries.
- Created a painting application using Canvas and similar technologies.
- Developed a bookstore application using Android SDK.
Software Engineer | Internship
Minimax Technologies
- Built a photo frame mobile application with multiple layouts, frames, templates, stickers, backgrounds, and text fonts to create incredible photos.
- Developed a voice recording mobile application with custom audio signal visualization.
- Created a flashlight mobile application with cool features like auto light on and off and timer.
Software Engineer | Internship
Havanour Technologies
- Developed tested, idiomatic, and documented websites using HTML, CSS, and JavaScript.
- Created multiple UI components of the website using HTML and CSS.
- Wrote and debugged code that would work across different browsers.
Experience
Con-Detect | Detecting Adversarially Perturbed Natural Language Inputs to Deep Classifiers
https://www.techrxiv.org/articles/preprint/Con-Detect_Detecting_Adversarially_Perturbed_Natural_Language_Inputs_to_Deep_Classifiers_Through_Holistic_Analysis/19295534We introduce an unsupervised detection methodology for detecting adversarial inputs to NLP classifiers. We note that minimally perturbing an input to change a model's output, a significant strength of adversarial attacks, is a weakness that leaves unique statistical marks reflected in the cumulative contribution scores of the input. Particularly, we show that the cumulative contribution score, called the CF-score of adversarial inputs, is generally greater than that of the clean inputs. We thus propose Con-Detect, a contribution-based detection method for detecting adversarial attacks against NLP classifiers. Con-Detect can be deployed with any classifier without having to retrain it. We show that it can reduce the attack success rate (ASR) of different attacks from 100% to as low as 0% for the best cases and =70% for the worst case. Even in the worst case, we note a 100% increase in the required number of queries and a 50% increase in the number of words perturbed, suggesting that Con-Detect is hard to evade.
Tamp-X | Attacking Explainable Natural Language Classifiers Through Tampered Activations
https://www.sciencedirect.com/science/article/pii/S0167404822001857We proposed first-of-its-kind Tamp-X, a novel attack that tampers the activations of robust NLP classifiers forcing the state-of-the-art white-box and black-box XAI methods to generate misrepresented explanations. Through extensive experimentation, we show that the explanations generated for the tampered classifiers are unreliable and significantly disagree with those generated for the untampered classifiers, even though the output decisions of tampered and untampered classifiers are almost always the same. Additionally, we study the adversarial robustness of the tampered NLP classifiers and find out that the tampered classifiers, which are harder to explain for the XAI methods, are also harder to attack by adversarial attackers.
All Your Fake Detector Belong to Us | Evaluating Adversarial Robustness of Fake-news Detectors
https://ieeexplore.ieee.org/abstract/document/9446139Additionally, we explore how changing the detector complexity, the input sequence length, and the training loss affects the robustness of the learned model. Our experiments suggest that RNNs are robust compared to other architectures and our evaluations provide vital insights to robustify fake-news detectors against adversarial attacks.
The Art in Our Worlds
https://github.com/msulemannkhan/nasaspaceappWe developed a mobile application using ML/AI techniques that allows users to input short text phrases, match that input to NASA science data or imagery, and display the results for the user creatively and artistically.
Domain Adaptation for Emotion Detection from Face Expressions
https://github.com/msulemannkhan/msds19011_Project_DLSpring2020Deep learning algorithms are efficient for facial expression classification, but these algorithms demand a high amount of data. Domain Adaptation can be used to address the lack of sufficient data. Right Now, we don't have much data on Pakistani facial expressions. In this Project, we created data set of Pakistani facial expressions and used domain adaptation to develop an efficient facial expression algorithm of Pakistani faces. We have achieved 58% accuracy with a baseline of 32 %.
AI Article Writer
https://github.com/msulemannkhan/text-generationI developed the project to demonstrate the capabilities of state-of-the-art NLP models. This demonstrates how we can use GPT Neo and GPT-3 to write blogs.
Bookstore Application
The application features a user-friendly interface that allows users to easily search for books by title, author, or genre. Users can add books to their shopping cart and proceed to checkout, where they can enter their payment information and complete their purchases.
The back end of the application is built using Django, which allows for efficient data management and user authentication. The application uses SQL and Postgres to store and manage user and product data, ensuring data security and scalability.
Overall, the Bookstore Application is a versatile and user-friendly platform that enables the local bookstore to sell their books online, providing customers with a convenient way to shop and support their favorite local bookstore.
Education
Master's Degree in Data Science
Information Technology University - Lahore, Pakistan
Bachelor's Degree in Computer Science
Government College University - Lahore, Pakistan
Certifications
Build Text Classification Model with AWS Glue and Amazon SageMaker
AWS
Introduction to Amazon Kinesis Analytics
Learning Amazon Web Services Lambda
Learning Amazon SageMaker
Learning AWS CloudFormation
Efficient Python Production Workflows
Data Science Foundations | Python Scientific Stack
Artificial Intelligence Foundations | Machine Learning
Apache PySpark by Example
Amazon Web Services | Data Services
Advanced Pandas
AWS for Developers | Data-driven Serverless Applications with Kinesis
AWS Machine Learning | Building an Expense Tracker Using Amazon Textract
Skills
Languages
Python, SQL, Python 3, JavaScript, HTML, CSS, Java, SCSS, GraphQL
Libraries/APIs
TensorFlow, Pandas, NumPy, Matplotlib, Scikit-learn, PySpark, Folium, Shapely, PyTorch, React, Volley, Volley Android Library, Retrofit, Retrofit 2, jQuery, OpenCV, Flask-RESTful, Django ORM, Azure Blob Storage API, Keras
Tools
Seaborn, PyPI, Amazon SageMaker, AWS IAM, BigQuery, Amazon Athena, AWS CloudFormation, AWS Glue, Jupyter, Celery, Azure Machine Learning, Tableau, Apache Airflow, ChatGPT
Paradigms
Data Science, REST, Test-driven Development (TDD)
Platforms
Amazon EC2, Amazon Web Services (AWS), Docker, AWS Lambda, Jupyter Notebook, Twilio, Kubernetes
Storage
PostgreSQL, Amazon S3 (AWS S3), Databases, Data Pipelines, NoSQL, Data Lakes, Amazon Aurora, Azure Blobs, Redis
Other
Back-end, Artificial Intelligence (AI), Machine Learning, Natural Language Processing (NLP), FastAPI, Data Visualization, Regression, Linear Regression, Data Analytics, Deep Learning, GPT, Generative Pre-trained Transformers (GPT), Data Wrangling, API Gateways, Google BigQuery, Machine Learning Operations (MLOps), Amazon Kinesis, Serverless, Data Analysis, Data Warehousing, Gunicorn, CI/CD Pipelines, Identity & Access Management (IAM), Infrastructure as Code (IaC), Big Data, Neural Networks, Deep Neural Networks, Recurrent Neural Networks (RNNs), Convolutional Neural Networks (CNN), Adversarial Attacks, Explainable Artificial Intelligence (XAI), BERT, Computer Vision, APIs, Google Colaboratory (Colab), OpenAI, Generative Pre-trained Transformer 3 (GPT-3), Text Generation, Web Development, Data Transcription, AI Design, Audio, Speech to Text, Speech Recognition, Text to Speech (TTS), Leadership, Architecture, Security, Data Engineering, Prefect, Azure Databricks, Ray.io, Cloud Computing, Containerization, Ray Train, Ray Tune, Ray Serve, Ray Core, Teamwork, Supervised Machine Learning, Supervised Learning, Classifier Development
Frameworks
Hadoop, Serverless Framework, Android SDK, Angular, Dagger, Bootstrap, Flask, Django, Selenium
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring