https://saeidaliei.github.io/posts/physical-intuitions/Sun, 16 Aug 2026 00:00:00 +0000https://saeidaliei.github.io/posts/physical-intuitions/Currently Large Language Models (LLM) coupled with Rienforcement Learning with Verifiable Rewards (RLVR) are winning math olympiads, proving theorems etc. which is to be expected, because you have Lean in math, but not the equivallent in physics. I suspect at least in short-term LLMs are unable to gain “true” physical intuition, at least not in the sense that giants like Einstein had. Nonetheless this is a collection of physics arguments framed around the reasoning that makes a strange result feel inevitable once you see it the “right way”.https://saeidaliei.github.io/posts/consciousness-computability-decidability-evolvability/Sat, 15 Aug 2026 00:00:00 +0000https://saeidaliei.github.io/posts/consciousness-computability-decidability-evolvability/1. The starting intuition, and why it’s too strong as stated # Gödel’s incompleteness theorems show that any sufficiently expressive, consistent formal system \$F\$ contains a true statement \$G\$ it cannot prove from its own axioms — informally, $$ G \leftrightarrow \neg \mathrm{Provable}_F(G). $$ It’s tempting to reach for an analogy: what if consciousness is a real physical phenomenon whose complete truth can’t be derived from whatever formal or mathematical framework we use to describe physical reality?https://saeidaliei.github.io/posts/terrarium-evolvability-lab/Sat, 15 Aug 2026 00:00:00 +0000https://saeidaliei.github.io/posts/terrarium-evolvability-lab/This is a companion piece to Consciousness, Computability, Decidability, and Evolvability. That piece argued for attacking weaker, well-formed questions about consciousness-adjacent proxies instead of the metaphysical question directly, and singled out evolvability as the more empirically checkable of the two tracks, since a negative evolvability result can be tested against biology in a way a negative decidability result cannot. This piece works through what that actually looks like as a formal object, and what a controllable synthetic world buys you when trying to study it.https://saeidaliei.github.io/posts/working-with-large-language-models/Sat, 15 Feb 2025 00:00:00 +0000https://saeidaliei.github.io/posts/working-with-large-language-models/Large language models are experiencing their Cambrian explosion. They may not be the path to AGI, but at least they give a taste of what it could be. The current mainstream approach to rely on scaling, may not be the sole solution, as the data runs out, and the models are plateauing in benchmarks, though we may very well see emergent behaviors that surprise us, as the bitter lesson taught us.https://saeidaliei.github.io/posts/simd-programming-and-vector-optimizations/Mon, 03 Feb 2025 00:00:00 +0000https://saeidaliei.github.io/posts/simd-programming-and-vector-optimizations/Essentially all modern processors have the capability to apply instructions on a vector in one processing unit cycle instead of operating on a single scalar. Language designers and compiler developers have been trying hard to leverage these hardware capabilities by compiling scalar programs into vector instructions. One possible approach is using SIMD (Single Instruction, Multiple Data) intrinsics, supported by all modern C/C++ compilers, through SSE (Streaming SIMD Extension), AVX (Advanced Vector Extensions) others for x86 architectures, and ARM NEON extensions.https://saeidaliei.github.io/posts/tiled-matrix-multiplication/Sat, 25 Jan 2025 00:00:00 +0000https://saeidaliei.github.io/posts/tiled-matrix-multiplication/The idea is simple, reuse the data already loaded into the cache as much as possible. Tiled algorithms, in general, divide the problem into smaller, manageable tiles that fit into faster, but limited-size memory, be it cache, shared memory, or registers, to improve memory access patterns. Some examples of tiled algorithms include: Matrix Multiplication: Divides the input matrices into submatrices (tiles), and computes partial multiplications for each tile. It uses shared memory in GPUs to store tiles, reducing global memory accesses.https://saeidaliei.github.io/posts/world-models/Sun, 19 Jan 2025 00:00:00 +0000https://saeidaliei.github.io/posts/world-models/At CES 2025, NVIDIA announced several interesting projects, including a single chip of 72 Blackwell GPUs, with 1.5 ExaFLOPS FP4 performance! Among the announcements was its first world foundation model, Cosmos. DeepMind is working on a similar model, and this is also one of xAI’s core missions. I suspect that these models are among the next frontier in scientific and AI research, following AlphaFold’s breakthroughs in protein folding prediction, and now Microsoft’s MatterGen model, which is specifically tailored for material science.https://saeidaliei.github.io/posts/spack-package-manager/Tue, 14 Jan 2025 00:00:00 +0000https://saeidaliei.github.io/posts/spack-package-manager/Recently I gave a talk/mini workshop on spack package manager at SKAO (available here), for which I dug into the core of spack. Spack boasts itself as the package manager for HPC, and it really is. The idea is straightforward, streamline what we normally do when compiling and installing software manually on HPC systems using Python. It’s a simple idea but not at all easy to implement. Compiling software is not an easy task, especially the scientific software and the ones that are designed to run on HPC systems on a large scale.https://saeidaliei.github.io/about/Mon, 01 Jan 0001 00:00:00 +0000https://saeidaliei.github.io/about/Hi! # I’m Saeid, a software engineer at the Square Kilometre Array Observatory (SKAO) headquarters near Manchester, UK. SKAO is set out to build the world’s largest radio telescope, with interferometry between many dishes and antennas in South Africa and Australia. Before joining SKAO, I was an AI/HPC engineer at the Institute of Research and Technological Innovation (RIT) at Area Science Park in Trieste, Italy, while being a research assistance professor at the College of Science and Technology (CST) of Temple University in Philadelphia, USA, which the H1B Visa process for never went through.https://saeidaliei.github.io/blog/Mon, 01 Jan 0001 00:00:00 +0000https://saeidaliei.github.io/blog/