Esmaeil Mirvakili
Graduate Researcher LLM Research
United States
Summary
👤
Senior🎓
Top SchoolEsmaeil Mirvakili is an applied AI engineer and PhD researcher specializing in long-horizon LLM agents, with ten years of experience building verifiable, deployable systems that prioritize reproducible evidence over flashy demos. At UC Santa Cruz he boosted agent accuracy by 15.2 F1 while cutting context tokens ~70%, created a closed-world benchmark for execution-based agent evaluation, and shipped a Kubernetes-distributed LLM training stack enabling 32-GPU jobs with end-to-end observability. His background spans full-stack and backend engineering—contributions include iOS work on a popular React Native audio recorder and production ML systems that reduced Ceph tail latency by ~40%. Esmaeil blends rigorous ablation-driven research with pragmatic engineering: benchmarks, eval harnesses, and scalable training/serving rails are integral parts of his products, not afterthoughts. He is based in the United States and combines academic depth with a track record of shipping reliable, observable infrastructure for real-world ML deployment.
10 years of coding experience
4 years of employment as a software developer
University of California Santa CruzMaster of Science - MS, Software Engineering, Master of Science - MS, Software Engineering at Sharif University of Technology
Stackoverflow
588reputation
134kreached
30answers
2questions
Github Skills (12)
objective-c10
ios10
react-native10
base64-encoding8
xcode6
html6
css6
twitter-bootstrap6
swift6
javascript6
uiscrollview6
iphone6
Programming languages (10)
Github contributions (5)
jsierles/react-native-audio
Apr 2018 - May 2018
Audio recorder library for React Native
Role in this project:
Mobile Developer (iOS)
Mobile Developer (iOS)Contributions:8 commits, 7 PRs in 1 month
Contributions summary:Esmaeil primarily contributed to the iOS side of the React Native audio recorder library. They fixed a typo, addressed an error that could occur when base64 encoding was disabled, and added functionality to return the audio file size and peak metering values. Furthermore, the user modified the example component to utilize the new file size parameter in the iOS environment.
reactreact-nativeaudioaudio-recorderrecorder
liamse/react-native-call-observer
Jun 2018 - Jun 2019
react-native module to get the status of calls with CXCallObserver
Contributions:22 commits, 1 PR, 16 pushes in 1 year
reactreact-native-modulecallsreact-nativestatus
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.