Skip to content
View toilaluan's full-sized avatar

Highlights

  • Pro

Block or report toilaluan

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
toilaluan/README.md

Tran Thanh Luan - toilaluan

I'm Luan, a Senior AI Engineer and Tech Entrepreneur with expertise in Machine Learning, Deep Learning, Generative Models, Computer Science, and System Design.

I'm the founder of condenses.ai, a startup providing prompt compression services built on a Decentralized Network (Subnet 47 - Bittensor). I developed the validation framework that incentivizes miners to contribute their best compression algorithms, creating an AI competition where performance determines rewards. At its peak, this subnet captured nearly 1% of network emissions. While condenses.ai is technically functional, we're still working toward market adoption and revenue generation.

I graduated from Hanoi University of Science and Technology with a Bachelor's in Computer Science. My professional experience includes:

  • Head of Decentralized AI at ZenAI, I lead the team to build subnets on Bittensor with various incentive mechanisms: decentralized image & llm inference (subnet 23), decentralized contest host (subnet 61), decentralized api bridge (subnet 18)...
  • Senior AI Engineer at ZenAI, where I worked on Generative Models for image, video, and text personalization, fine-tuning diffusion models, and customizing ComfyUI pipelines.
  • Data Scientist at Pixta Vietnam, where I built an Auto Review System for Pixta Stock and distilled LLMs for specific domains

Beyond my professional projects, I'm committed to advancing my AI expertise through independent research and experimentation. I regularly work on training models from scratch (LLM, Diffusion, Flow Matching related models), learning to optimize CUDA processing, Triton-lang, and implementing state-of-the-art papers. My versatility allows me to excel in both cutting-edge research and practical engineering applications across machine learning, deep learning, and broader computer science fields.

Pinned Loading

  1. latent-llm latent-llm Public

    Python

  2. simple-flux-trainer simple-flux-trainer Public

    Python

  3. SocialTensor/SocialTensorSubnet SocialTensor/SocialTensorSubnet Public

    Python 22 36

  4. LogicNet-Subnet/LogicNet LogicNet-Subnet/LogicNet Public

    Forked from SocialTensor/SocialTensorSubnet

    Supercharge Bittensor Ecosystem with Advanced Mathematical and Logical AI

    Python 14 13

  5. condenses/neural-condense-subnet condenses/neural-condense-subnet Public

    1 4

  6. corcel-api/cortex.t corcel-api/cortex.t Public

    Subnet 18

    Python 1 3