Practical ML engineering notes. Deep dives for engineers and accessible explainers for everyone.https://theartificialengineer.ai/en-ushttps://theartificialengineer.ai/posts/iclr-2026-ml-architectures-orals/https://theartificialengineer.ai/posts/iclr-2026-ml-architectures-orals/Notes from Day 2's first oral session at ICLR 2026. Five papers on learning rate schedules, MoE sparsity, curriculum data, softmax expressivity, and pre-training under infinite compute.Fri, 24 Apr 2026 00:00:00 GMTengineersiclrconferencepretrainingmoescaling-lawscurriculumtransformershttps://theartificialengineer.ai/posts/iclr-2026-llm-evaluation-orals/https://theartificialengineer.ai/posts/iclr-2026-llm-evaluation-orals/Notes from the second oral session at ICLR 2026. Six papers on reward hacking detection, multi-turn reliability (the Best Paper), micro-benchmarking, value-difference generation, preference interpretability, and mutual-judgment alignment.Thu, 23 Apr 2026 00:00:00 GMTengineersiclrconferencellmsevaluationalignmentrlhfhttps://theartificialengineer.ai/posts/iclr-2026-llm-reasoning-orals/https://theartificialengineer.ai/posts/iclr-2026-llm-reasoning-orals/Notes from the first oral session at ICLR 2026. Five papers on active reasoning, long-context memory, CoT verification, retriever training, and model merging.Thu, 23 Apr 2026 00:00:00 GMTengineersiclrconferencellmsreasoningrllong-contextretrievalinterpretabilityhttps://theartificialengineer.ai/posts/iclr-2026-mataric-human-centered-ai/https://theartificialengineer.ai/posts/iclr-2026-mataric-human-centered-ai/Opening keynote at ICLR 2026 from USC and Google DeepMind's Maja Mataric — a case that intelligence without a body misses the point, and that the robots we actually need aren't the ones trying hardest to look human.Thu, 23 Apr 2026 00:00:00 GMTeveryoneiclrconferencehuman-centered-airoboticsembodimenthttps://theartificialengineer.ai/posts/why-this-blog/https://theartificialengineer.ai/posts/why-this-blog/A two-track ML blog: reproducible engineering deep dives for practitioners, and clear explainers for everyone.Tue, 02 Sep 2025 00:00:00 GMTeveryoneml-engineeringreproducibilitylearning-in-public