Abdullah M. is a Machine Learning Engineer at DPD UK. Colleagues describe them as "It's a privilege to write this recommendation for Abdullah Mujeeb, who worked under my supervision as a Senior Data Scientist in the Data Science Team at Daraz. Abdullah was a valuable member of our team for 2 and a half years and made exceptional contributions to the company. As a Senior Data Scientist, Abdullah was responsible for working on the recommender system team and made ground-breaking discoveries. He was a skilled data scientist and was able to train the team and deliver impactful projects that had a significant impact on the business. He was also a valuable asset to the team, contributing his knowledge and expertise to the advancement of our data science initiatives. Abdullah was a dedicated and hard-working employee who consistently exceeded expectations. He was a quick learner and was always willing to take on new challenges. He was an excellent communicator and was able to collaborate effectively with others to solve complex problems. During his tenure, Abdullah demonstrated his ability to manage projects from start to finish and was always willing to go the extra mile to ensure that projects were delivered on time and to the highest standards. He was an effective leader and was able to motivate the team to achieve their goals. I am confident that Abdullah will be an asset to any organization he joins. He has a bright future ahead of him, and I would not hesitate to recommend him for any future endeavors." and "Abdullah is one of those members in the team who are unforgettable. His passion towards his work was evident, not only in his drive to continually grow technically, but also taking care to understand the business impacts and view things from a broader angle. Furthermore, he was always willing to lend a helping hand to colleagues, as well as share insights into the latest research paper he was reading. Definitely will contribute positivity and positively to any team which will be lucky to have him!"
-
Machine Learning EngineerDpd UkLondon, England, Gb -
Machine Learning EngineerDpd Uk Nov 2023 - PresentOldbury, England, United KingdomForecasting the future :D -
Lead Data ScientistDastgyr Jun 2023 - Nov 2023Remote● Implemented SWING (https://arxiv.org/abs/2010.05525) from scratch to improve recommendations on Top For You module (Hitrate went from 5% to 18.94%)● Currently researching best MLOps platforms to setup logging, monitoring, AB Testing and model scheduling including inference in realtime [neptune, sagemaker etc]● Enforced culture of documenting everything. -
Session LeadUdacity Sep 2022 - Apr 2023Session Lead for AI Programming with Python -
Senior Data ScientistAlibaba Group Jun 2022 - Feb 2023Karāchi, Sindh, Pakistan● [Algo] Implemented Daraz’s first deep model ‘DeepFM’ for ranking product assortment for users – this resulted in an increase of 5% uplift in click through rate over baseline logistic regression. ● [Algo] Introduced Re-Rank stage in Daraz to provide operators with a more aggressive control over [click, add-2-cart, order] by implementing LTR. This resulted in an increase of ~34% conversion rate over baseline. ● [Algo]Collaborated with Search division for Query2Query to remove bad rewrites and observed an uplift in search conversion rate by 2% [w.r.t query rewrites] ● [Product] Developed Trending Now backend for recommendation and recall strategy, two modules of Trending Now were A/B tested on the product and the best variant showed multiple categories with ~20% CTR, ~0.3% CR. [+26% CR, 46% CTR than Single Category Trending Now] ● [Product] Developed the new dMart’s backend for recommendation and supported operators understand how to exploit the algorithm correctly. Over the base bucket (hot items only vs personalized), CTR +30%, CR 100%, and ranking with LR enabled us to take this even further over the personalized bucket [~13% CR, ~3% CTR] ● [Product] Introduced Search Tips in Search using the recall techniques from recommendation for refining user query or recommending entirely a new one, which uplifted overall Search CR by ~5%. ● [Optimization] Optimized backup logic algorithm by utilizing feature importance and cutting costs in compute by a significant margin. [daily schedulable over weekly schedule] ● [Experimentation] Implemented Reliable Negative Sampling to handle class imbalance which resulted from 1:50 to 1:10 imbalance. ● [Algo]Reduced item-cold start by introducing the only content-based algorithm under our sleeves to witness 0 view products gaining limelight and 2nd highest CTR in recall phase. Due to low coverage, a multi-recall in one strategy is underway. -
Data ScientistAlibaba Group Oct 2020 - Jun 2022Karāchi, Sindh, Pakistan -
Teaching AssistantNational University Of Computer And Emerging Sciences Feb 2020 - Jul 2020Karāchi, Sindh, PakistanAssisting the lecturer with vivas, quizzes and projects in Computer Networks -
Seo Content WriterCrea8Ive Bench Jun 2018 - Jul 2018Karāchi, Sindh, Pakistan- Writing blogs, articles with Keywords to maximize reach- Nearly 100 blogs written within 2 months time-frame
Abdullah M. Education Details
-
Data Science -
Computer Science -
Defense Authority Sheikh Khalifa Bin Zayed CollegePre-Engineering -
Jawahir Al RiyadhScience
Frequently Asked Questions about Abdullah M.
What company does Abdullah M. work for?
Abdullah M. works for Dpd Uk
What is Abdullah M.'s role at the current company?
Abdullah M.'s current role is Machine Learning Engineer.
What schools did Abdullah M. attend?
Abdullah M. attended University Of Essex, National University Of Computer And Emerging Sciences, Defense Authority Sheikh Khalifa Bin Zayed College, Jawahir Al Riyadh.
Who are Abdullah M.'s colleagues?
Abdullah M.'s colleagues are Daniel Shone, Chris Rhodes, Hannah Price, Mark Woodcock, Kaloyan Ivanov, Megan Mcdonagh, Nicolae Dumitru.
Not the Abdullah M. you were looking for?
-
-
Abdullah Ahmad Kamal, ACCA, CA (M)
London Area, United Kingdom -
Abdullah M.
Cardiff -
-
Abdullah M.
London
Free Chrome Extension
Find emails, phones & company data instantly
Aero Online
Your AI prospecting assistant
Select data to include:
0 records × $0.02 per record
Download 750 million emails and 100 million phone numbers
Access emails and phone numbers of over 750 million business users. Instantly download verified profiles using 20+ filters, including location, job title, company, function, and industry.
Start your free trial