Data science, career and other topics
27 October 2025
A biologically inspired LLM built as a graph of spiking neurons with Hebbian learning — it matches GPT-2 scaling while producing sparse, interpretable activations that map to clear concepts.
06 October 2025
My review of the paper LongLive Real-time Interactive Long Video Generation
28 September 2025
My experience of interview preparation as MLE in 2024-2025
15 September 2025
My review of the paper Sharing is Caring Efficient LM Post-Training with Collective RL Experience Sharing
01 September 2025
My review of the paper Pref-GRPO Pairwise Preference Reward-based GRPO for Stable Text-to-Image Reinforcement Learning
25 August 2025
Meta's self-supervised vision model trained on 17 billion images, introducing Gram anchoring to prevent feature degradation at scale and achieving state-of-the-art across diverse vision tasks without fine-tuning.
