Chanchana Sornsoontorn
Senior AI ML Algorithm Developer at Six Atomic | AI Solutions for Real-Time Fashion
Bangkok, Thailand
Summary
🤩
Rockstar🎓
Top SchoolChanchana Sornsoontorn is a Senior AI/ML Algorithm Developer based in Bangkok with 10 years of experience building production-ready models, from LLM chatbots with RAG and LangChain to generative image models like Stable Diffusion and CNN classifiers. He combines a first-principles, systems-thinking approach—having worked through Thinking in Systems exercises—with a competitive programming background that sharpens his algorithmic rigor. At SolidTen.ai he built an end-to-end AI image generator and invented market-differentiating features by reimplementing research ideas from scratch, and at Six Atomic he leads real-time fashion AI solutions and hands-on ML research. His open-source contributions focus on improving educational and tooling repos (notably coursework for a popular CADL MOOC and practical Kalman filter notebooks), highlighting a talent for clarifying complex concepts. Colleagues note he often spots strategic dead ends months earlier than others, making him both a pragmatic implementer and a strategic thinker.
9 years of coding experience
Phanomsarakham "Phanom Adun Witthaya"Bachelor of Applied Science (BASc), Computer Science, Bachelor of Applied Science (BASc), Computer Science at Burapha University
English, Thai
Stackoverflow
3,947reputation
3.2mreached
33answers
41questions
collections
top-5%
optimization
top-5%
Github Skills (21)
python10
machine-learning10
kalman-filter10
deep-learning10
tensorflow10
file-handling10
object-detection10
computer-vision10
jupyter-notebook10
command-line-interface9
command-line9
visual-studio9
optimization9
cli9
data-analysis9
Programming languages (20)
Github contributions (5)
Cartucho/mAP
May 2018 - May 2018
mean Average Precision - This code evaluates the performance of your neural net for object recognition.
Role in this project:
ML Engineer
ML EngineerContributions:10 commits, 2 PRs, 4 comments in 6 days
Contributions summary:Chanchana primarily focused on improving the maintainability and usability of the mAP evaluation tool. They implemented a script to handle files with inconsistent names and ensure a consistent ground truth and predicted data structure. They also added functionality to rename classes, remove spaces, and ensure file matching for object detection evaluation. Additionally, the user introduced a method to prevent file not found errors.
detectiondeep-learningobject-recognitionrecognitionobject-detection
pkmital/CADL
Sep 2016 - Jan 2017
Course materials/Homework materials for the FREE MOOC course on "Creative Applications of Deep Learning w/ Tensorflow" #CADL
Role in this project:ML Engineer
ML EngineerContributions:18 commits, 12 PRs, 11 comments in 3 months
Contributions summary:Chanchana primarily contributed to the course materials for a deep learning course. Their contributions involved fixing typos, correcting values, and improving code clarity in the Jupyter Notebooks, which included changes in code examples using TensorFlow. They also added hints and warnings to improve the learning experience for students. Furthermore, the user addressed issues within the code, such as fixing image sizes and the number of iterations.
deep-learningtensorflowmooc
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.