Alykhan Tejani

AI/ML Engineer

Reach out

I’m an AI/ML engineer with over 10 years of experience designing, training, and deploying large-scale machine learning models. I’ve led teams at startups and big tech companies, focusing on deep learning, distributed training, and AI research.

Most recently, I’ve been working on large-scale recommender systems at ShareChat, optimising ranking models and user engagement for one of India’s largest social platforms. My background includes contributions to open-source projects like PyTorch and publications with over 70,000 citations.

I’m passionate about building intelligent systems that scale, blending cutting-edge research with real-world impact.

About Me

Relevant
Experience

Developed and optimised content recommendation models for timeline ranking and ads.
Worked on large-scale distributed training, improving efficiency and scalability.
Designed deep learning models for engagement prediction, personalised recommendations,
and relevance ranking.
Contributed to infrastructure improvements, accelerating model iteration cycles for
production ML.
worked on computer vision super-resolution research, improving image quality for media
compression, enhancement, and content delivery.

Staff Machine Learning Engineer

Twitter

2016 - 2022

Leading the development of large-scale recommender systems, optimising ranking models for
personalised content discovery. [blog post]
Designed and deployed models to enhance content recommendations, user engagement, and
retention.
Scaled distributed training and inference for millions of users, improving model efficiency and
serving latency.
Worked on multi-modal embeddings, representation learning, and dynamic user modelling to
improve content relevance.
Focused on productionising AI at scale, collaborating across teams to refine feature
engineering, model serving, and experimentation.

Senior staff ML Engineer

Sharechat/Moj

2022 - Present

Co-developed the augmented reality tracking engine for mobile devices, powering Blippar’s
AR app.
Led deep learning research for large-scale visual object recognition, enabling Blippar’s visual
browser.
Work featured in major media outlets including The Daily Mail, The Telegraph, The Wall
Street Journal, and demoed on Bloomberg, CNBC, Fox News NY, and BBC London News.

UK Lead research engineer

BlippAR

2014 - 2016

I am a firm believer in the power of open-source to drive innovation and push the boundaries of machine
learning. I created Ignite, a high-level library for training neural networks, designed to simplify deep learning
workflows. I have also contributed extensively to PyTorch, with commits improving its core functionality, and
was previously a maintainer of torchvision, helping to enhance one of the most widely used libraries for
computer vision research.

My work in open source has been driven by a passion for scalable, efficient ML systems, and I continue to
support the community through contributions, mentoring, and sharing research-driven insights.

Open Source

torchvision

Publications

D. Guo, S. I. Ktena, F. Huszar, P. K. Myana, W. Shi, A. Tejani

RecSys 2020

[PDF]

Deep Bayesian bandits: Exploring in online personalized recommendations

C. Zhang, Y. Liu, Y. Xie, S. I. Ktena, A. Tejani , A. Gupta, P. K. Myana, D. Dilipkumar, S. Paul, I. Ihara, P.
Upadhyaya, F. Huszar, W. Shi

RecSys 2020

[PDF][Video]

Model Size Reduction Using Frequency Based Double Hashing for
Recommender Systems

L. Belli, S. I. Ktena, A. Tejani , A. Lung-Yut-Fon, F. Portman, X. Zhu, Y. Xie, A. Gupta, M. Bronstein, A. Delić,
G. Sottocornola, W. Anelli, N. Andrade, J. Smith, W. Shi

RecSys 2020

[PDF]

Privacy-Preserving Recommender Systems Challenge on Twitter’s Home
Timeline

B. Steiner, Z. DeVito, S. Chintala, S. Gross, A. Paszke, F. Massa, A. Lerer, G. Chanan, Z. Lin, E. Yang, A.
Desmaison, A. Tejani , A, Kopf, J. Bradbury, L. Antiga, M, Raison, N, Gimlelshein, S. Chilamkurthy, T. Killeen,
L. Fang, J. Bai

Neural Information Processing Systems (NeurIPS), 2019