John Leimgruber
Philadelphia, Pennsylvania, United States
Summary
🤩
Rockstar🎓
Top SchoolJohn Leimgruber is a seasoned systems and data engineer with 13 years of experience building reliable, cost-conscious infrastructure and data pipelines across embedded robotics, cloud hosting, and media SaaS. He combines low-level systems expertise—kernel compilation, eBPF performance tuning, and NUMA/cgroups analysis—with high-level data engineering and async Python services to expose predictable, typed APIs for complex legacy schemas. At Linode he sped provisioning and performance work that contributed to a major acquisition, and at Futurestay he both rescued legacy systems and cut cloud spend by $10k+/month while mentoring junior engineers. His background in robotics and embedded real-time systems informs pragmatic, performance-first designs, and he’s contributed ML-focused work to notable open-source projects like a char-RNN TensorFlow repo. Based in Philadelphia, he balances production engineering with creative side projects—from game mods to a published book—demonstrating a blend of technical depth and curious, hands-on inventiveness.
13 years of coding experience
17 years of employment as a software developer
MSECE, Electrical and Computer Engineering, MSECE, Electrical and Computer Engineering at Purdue University
Stackoverflow
26reputation
843reached
1answer
0questions
Github Skills (9)
rnn-model10
tensorflow10
python10
n10
machine-learning9
lstm9
dropout8
horizon6
rethinkdb6
Programming languages (17)
Github contributions (5)
sherjilozair/char-rnn-tensorflow
Mar 2017 - Apr 2017
Multi-layer Recurrent Neural Networks (LSTM, RNN) for character-level language models in Python using Tensorflow
Role in this project:
ML Engineer
ML EngineerContributions:23 commits, 18 PRs, 10 pushes in 1 month
Contributions summary:John contributed to the character-level language model by merging various branches, including patches and updates. Their work includes adding new functionality, such as a NASCell, updating MultiRNNCell, and integrating tensorboard for improved monitoring. They also addressed printing errors by encoding results and updated the sampling process.
language-modellstmpythonrecurrent-neural-networkstensorflow
ubergarm/dodocon
May 2014 - Dec 2015
Contributions:3 commits, 2 PRs, 2 pushes in 1 year 6 months
consul-clusterdockerkubernetesconsulcluster
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.