Salman Maqbool

Senior Machine Learning Engineer

PackageX

Jul 2023 - Current (2 years 9 months)

• Trained and deployed text and multimodal Transformer models such as BERT and LayoutLMv3 for iOS and Android • Trained and deployed vision models (YOLOv8) for detection and segmentation on Android and iOS • Defined metrics and designed and implemented a robust client-server-based end-to-end evaluation scheme for our mobile ML solution • Reduced ML model inference time by 60 % via quantization and various optimization techniques • Improved ML model accuracy by an absolute 20 % via targeted data annotation, data pre and post-processing, improving OCR quality, and better models • Developed and deployed an application for vehicle and license plate recognition

P

Founder

PSYCHIATRAI

Oct 2022 - Current (3 years 6 months)

• Leading the development of a fullstack mental health platform for patients, practitioners, and organizations: Waitlisted 25 clients in 3 days (and counting) • Developed and piloted a 6-week online intervention with both sync and async support for mild-moderate depression: Onboarded 15 clients for the intervention, with positive feedback and improvements in PHQ-9 scores

O

Technical and Regional Expert

OverCome

Apr 2022 - Oct 2022 (7 months)

• Partnered with Clear Global to translate Cogito and MindEase into Ukrainian for conflict affectees • Developed a study exploring the potential for smartphone data as a predictor of mental health • Designed and piloted a peer-guided behavioral change intervention developed using an agile approach - with positive client feedback • Developed a digital Cognitive Behavioral Therapy (CBT) prototype for anxiety and depression • Contributed to organizational strategy

L

AI Developer

Luxolis

Oct 2021 - Apr 2022 (7 months)

• Worked closely with mobile developers to develop data acquisition and client-side 3D model library apps (launched on App Store and Play Store) • Architected, developed and optimized a modular backend to receive 3D data from mobile apps and run object, scene, and human body reconstruction pipelines to generate 3D models • Introduced Agile and GitOps to the team for better development practices and developer productivity • My work directly contributed to the company getting a contract from KT Corporation

V

Machine Learning Engineer

VEEVE

Nov 2019 - Jul 2020 (9 months)

• Incorporated Destruction and Construction Learning (DCL) training paradigm to improve fine-grained grocery classification accuracy • Defined different evaluation metrics and implemented those so that we can continuously monitor our performance • Enhanced code readability and performance by redesigning and implementing it as GStreamer modules, optimizing deep learning models using DeepStream and TensorRT

P

Founder

PSYCHIATRAI

Oct 2019 - Mar 2021 (1 year 6 months)

• Carried out a literature review and market research to come up with a hypothesis for an intervention for mental health: Low-cost, minimal guidance in digitally delivered Cognitive Behavioral Therapy (CBT) can provide more psychologically- and cost-effective treatment • Designed a pilot study to test the hypothesis

C

ML + Python Engineer (TSP)

CERN, INSPIRE-HEP

Mar 2018 - Feb 2019 (1 year)

• Developed and deployed an LSTM research paper classifier with an accuracy of 91 %, reducing curation time and effort • Analyzed discrepancies in our reference matcher. Improved matching performance using ElasticSearch

C

Summer Student

CERN, ATLAS EXPERIMENT

Jun 2017 - Sep 2017 (4 months)

• Added a more robust track reconstruction algorithm (General Broken Lines) to the Proteus framework, improving track reconstruction accuracy

S

Design Engineer (ML)

SIMPLICITY LABS

Feb 2017 - May 2017 (4 months)

• Trained Convolutional Neural Networks (YOLO, ResNet) for fine-grained vehicle detection and classification on motorways, achieving 88 % accuracy over fine categories and 95 % accuracy over coarse categories

About

Detail

Timeline

Résumé