Muhammad Osama Tarar
Senior Machine Learning Engineer
Lahore
Summary
Astute Senior Machine Learning Engineer with data-driven and technology-focused approach. Communicates clearly with stakeholders and builds consensus around well-founded models. Talented in writing applications and reformulating models.
Overview
8
8
years of professional experience
15
15
years of post-secondary education
Work History
Senior Machine Learning Engineer
Senarios
Lahore
09.2023 - Current
- Expertise in Language Models (LLMs) and Natural Language Processing: Proficient in leveraging advanced Language Models (LLMs) such as GPT-3, Llama2, and Falcon to develop robust AI solutions, harnessing their sophisticated natural language processing capabilities.
- Innovative AI Solutions: Recognized for the successful design and deployment of cutting-edge AI solutions that integrate LLMs, effectively addressing diverse business needs spanning various industries.
- Collaborative Cross-functional Engagement: Demonstrated proficiency in seamlessly collaborating with cross-functional teams to discern precise business requirements and transform them into highly efficient AI-driven solutions.
- Continuous Research and Technological Proficiency: Dedicated to maintaining a leading edge in technological advancements by actively researching and staying updated with the latest developments in LLMs and computer vision.
- End-to-End AI Pipeline Implementation: A proven track record of executing end-to-end AI pipelines, encompassing data preprocessing, feature extraction, model training, and deployment, resulting in comprehensive and effective AI solutions.
- Industry Acumen and Best Practices: Keen on maintaining industry awareness and adopting best practices in AI engineering, continually enhancing the performance and functionality of AI systems.
PhD Student Research Assistant (Machine Learning)
LUMS
Lahore
09.2019 - Current
- Working on the prognostics and diagnostics of Lithium-ion batteries in the context of battery swapping stations
- Worked on the techno-economic framework for battery swapping stations, especially in the context of two- and three-wheelers in developing countries
- Also worked on estimating remaining useful life and degradation modes in Lithium-ion batteries using machine learning and deep learning architectures.
- Also worked on classification of lithium-ion cells using machine and deep learning architectures.
Research Assistant (Machine Learning)
LUMS
Lahore
09.2017 - 06.2019
- Worked on the recovery of finite rate of innovation signals, a class of non bandlimited signals having finite degrees of freedom
- The problem was that we have an ensemble of finite number of diracs on the sphere which are embedded in noise and the task is to estimate the amplitude and location of these diracs
- We divided the problem in two parts 1) Denoising, 2) Parameter Estimation.
- For denoising we employed deep learning architecture and for parameter estimation we used annihilating filter technique
- The work also resulted in my Masters Thesis titled as "Denoising and Parameter Estimation of an Ensemble of Finite Number of Diracs On the Sphere".
Research Assistant (Machine Learning)
LUMS
Lahore
08.2015 - 08.2017
- Carried out research over the recent work done in the field of ambulatory EEG and documented a report on the utility of ambulatory EEG and its implications in comparison to the conventional EEG method used
- Carried out research over different EEG data acquisition systems and carried out experiments with the Emotive and Open
- BCI to find out their utility of application in the medical field
- In the end, an extensive report was formulated discussing different types of acquisition systems available and their comparison was made
- Carried out research on the artifacts related to EEG and documented a report on these artifacts
- Also helped in the experimentation of classification of artifacts by providing the marked data of artifacts related to eye
- Also Visited Mayo Hospital to explore opportunity for collaboration
- Marked the CHB-MIT dataset and revised it 2-3 times for EEG artifacts to be reviewed by the technologist
- I also worked for data compression of EEG signals using coding techniques.
- Classification of epochs using machine learning algorithms.
Education
Ph.D. - Electrical Engineering
Lahore University of Management Sciences
Lahore, Pakistan 05.2019 - Current
Master of Science - Electrical Engineering
Lahore University of Management Sciences
Lahore, Pakistan 05.2017 - 05.2019
Bachelor of Science - Electrical Engineering
University of Engineering & Technology (UET)
Lahore, Pakistan 05.2011 - 05.2015
A-level - Pre Engineering
Garrison Academy For Boys
Lahore, Pakistan 09.2009 - 05.2011
O-level - Science
Garrison Academy For Boys
Lahore, Pakistan 09.2006 - 05.2009
Skills
Interpersonal Communication
undefinedPublications
Transactions/Journals:
[1] Muhammad Osama Tarar, Naveed UL Hassan, Ijaz Haider Naqvi, and Michael Pecht, “Techno-economic framework for electric vehicle battery swapping stations,” IEEE Transactions on Transportation Electrification, 2023.
[2] Muhammad Osama Tarar, Ijaz Haider Naqvi, Zubair Khalid, and Michal Pecht. "Accurate prediction of remaining useful life for lithium-ion battery using deep neural networks with memory features." Frontiers in Energy Research 11 (2023): 1059701.
[3] Shehla Amir, Moneeba Gulzar, Muhammad Osama Tarar, Ijaz Haider Naqvi, Nauman Ahmed Zaffar, and Michael Pecht, “Dynamic equivalent circuit model to estimate state-of-health of lithium-ion batteries,” IEEE Access, vol. 10, pp. 18279–18288, 2022.
Conference Publications:
[4] Muhammad Osama Tarar and Zubair Khalid. "Reconstruction of finite rate of innovation spherical signals in the presence of noise using deep learning architecture." In 2020 28th European Signal Processing Conference (EUSIPCO), pp. 1487-1491. IEEE, 2021.
[5] Muhammad Osama Tarar, Naveed Ul Hassan, and Ijaz Haider Naqvi, “On the economic feasibility of battery swapping model for rapid transport electrification,” in 2021 IEEE PES Innovative Smart Grid Technologies Europe (ISGT Europe), 2021, pp. 1–5.
[6] Muhammad Osama Tarar, Naveed UL Hassan, and Ijaz Haider Naqvi, “Modular approach towards battery swapping: Time and technical parameters quality tradeoff,” in 2021 IEEE PES Innovative Smart Grid Technologies - Asia (ISGT Asia), 2021, pp. 1–5.
[7] Muhammad Osama Tarar, Naveed Ul Hassan, and Ijaz Haider Naqvi, “Higher training size, increased model complexity or both: A novel decision framework for cycle life classification of lithium-ion cells,” in 2022 IEEE PES Innovative Smart Grid Technologies-Asia (ISGT Asia). IEEE, 2022, pp. 81–85.
Book Chapter:
[8] Nadeem Ahmad Khan, Gul Hameed Khan, Malik Anas Ahmad, M. Awais bin Altaf, and Muhammad Osama Tarar. "The Extended i-NSS: An Intelligent EEG Tool for Diagnosing and Managing Epilepsy." In International Joint Conference on Biomedical Engineering Systems and Technologies, pp. 243-262. Cham: Springer International Publishing, 2020.
Timeline
Senior Machine Learning Engineer
Senarios
09.2023 - Current
PhD Student Research Assistant (Machine Learning)
LUMS
09.2019 - Current
Ph.D. - Electrical Engineering
Lahore University of Management Sciences
05.2019 - Current
Research Assistant (Machine Learning)
LUMS
09.2017 - 06.2019
Master of Science - Electrical Engineering
Lahore University of Management Sciences
05.2017 - 05.2019
Research Assistant (Machine Learning)
LUMS
08.2015 - 08.2017
Bachelor of Science - Electrical Engineering
University of Engineering & Technology (UET)
05.2011 - 05.2015
A-level - Pre Engineering
Garrison Academy For Boys
09.2009 - 05.2011
O-level - Science
Garrison Academy For Boys
09.2006 - 05.2009
Similar Profiles
Ravail IrshadRavail Irshad
Technical Team Lead/Project Manager at SenariosTechnical Team Lead/Project Manager at Senarios
Farooq KhalidFarooq Khalid
Lead SQA Engineer at Exarta LabsLead SQA Engineer at Exarta Labs
Muhammad Arham NaseemiMuhammad Arham Naseemi
Art Director at FreelanceArt Director at Freelance