Data science, career and other topics
29 October 2026
A review of Jeva, a decision-native architecture for system one models that encodes the request and each option independently, refines only the request state against reusable option memory, and adds a separate coverage gate for null outcomes.
26 October 2026
A review of Beyond LLMs: The Next Frontier of AI by Sahir Maharaj, a practical book that builds, chapter by chapter, on one customer-support system to engineer everything around the language model (explicit state, memory, bounded agents, deterministic rules, evaluation, and cost).
05 October 2026
My advice for beginners in machine learning: learning the fundamentals, finishing projects, using AI tools, and preparing for a first job. An update to my 2018 data science roadmap.
21 September 2026
DeepSeek-V4.1-Flash is a 552B multimodal MoE model that attacks KV cache cost along all three of its multiplicative dimensions at once, then redesigns prefill and cache persistence around the result. It activates 8B parameters per token during prefill and holds 890 bytes of global cache per token.
10 September 2026
A write-up of the ROGII Wellbore Geology Prediction competition: how I adopted an AI driven approach and got a silver medal.
08 September 2026
Qwen-Drive-1.0 turns Qwen3.5-4B into a unified autonomous-driving model for 3D perception, scene understanding, reasoning, and trajectory planning while largely preserving its general vision-language capabilities.
