Kohulan Rajan
Project Lead For AI And Automation
Jena, Thuringia, Germany
Summary
👤
Senior🎓
Top SchoolKohulan Rajan is a project lead for AI and automation who turns cutting-edge computer vision and NLP research into production systems for cheminformatics, most notably as lead developer of DECIMER.ai — the first comprehensive open-source platform for automated chemical structure recognition published in Nature Communications and featured by Google Cloud. With 8+ years of experience and a PhD bridging chemistry and AI, he combines deep research (21+ papers, 1,600+ citations) with production engineering in TensorFlow/PyTorch, FastAPI, Docker/Kubernetes and GCP to scale ML from prototypes to global platforms. He co-developed the COCONUT natural products database, accelerating open drug-discovery workflows, and has been recognized as a Google TPU Research Cloud Fellow and CVPR demo presenter. Beyond code and papers, he mentors students, teaches AI/ML, champions open science, and brings a creative edge as a professional photographer and published Tamil poet — a background that informs his pragmatic yet inventive approach to problem solving.
8 years of coding experience
15 years of employment as a software developer
GCE A/Level, Biology, General, GCE A/Level, Biology, General at Highlands College Sri-lankaDoctor of Philosophy - PhD, Chemistry and AI, Doctor of Philosophy - PhD, Chemistry and AI at Friedrich Schiller University JenaDiploma in Animation, 3D animation, Video Editing, Special Effects, Diploma in Animation, 3D animation, Video Editing, Special Effects at Wijeya Graphics InstitutionsGCE O/Level, Junior High/Intermediate/Middle School Education and Teaching, GCE O/Level, Junior High/Intermediate/Middle School Education and Teaching at Holy Trinity Central CollegeMaster of Science (MSc), Bioinformatics, Master of Science (MSc), Bioinformatics at Savitribai Phule Pune UniversityBengaluru University
English, Tamil, Sinhala, Hindi, marati, German, Spanish
Github Skills (82)
transformers10
python10
deep-learning10
autoencoder10
chemistry10
translator10
comparator10
baidu10
pytorch10
neural-machine-translation10
image-recognition10
tpu10
bioinformatics10
encoder9
cheminformatics9
Programming languages (8)
Github contributions (5)
Kohulan/DECIMER-Image_Transformer
Sep 2020 - Dec 2022
DECIMER Image Transformer is a deep-learning-based tool designed for automated recognition of chemical structure images. Leveraging transformer architectures, the model converts chemical images into SMILES strings, enabling the digitization of chemical data from scanned documents, literature, and patents.
Contributions:12 releases, 7 reviews, 135 commits in 2 years 3 months
pytorchtransformersimage-recognitiondeep-learningtpu
Kohulan/DECIMER-Image-to-SMILES
May 2020 - Sep 2021
The repository contains the network and the related scripts for auto-encoder based Chemical Image Recognition
Contributions:1 release, 62 commits, 1 PR in 1 year 3 months
autoencoderpythonimage-recognitiongenerationchemical
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.