Ayoub Ben Chaliah

AdaLLM is an NVFP4-first inference runtime for Ada Lovelace (RTX 4090) with FP8 KV cache and custom decode kernels. This repo targets NVFP4 weights and keeps the entire decode path in FP8

a novel architecture that leverages Autoencoders to superimpose the hidden representations of a base model and a fine-tuned model within a shared parameter space. Using B-spline-based blending coefficients and autoencoders that adaptively reconstruct the original hidden states based on the input data distribution.

TensaLang is a Tensor-first programming language, compiler, and runtime that let you write the Model’s inference engine (e.g. LLMs) and sampling in high level language, then compile it through MLIR to Multiple targets (e.g. CPU, CUDA, ROCm)

GANs for Time series analysis (Synthetic data generation, anomaly detection and interpolation), Hypertuning using Optuna, MLFlow and Databricks

Many market analysts believe that predicting market’s stocks fluctuations is nearly impossible to achieve due to the number of variables involved, especially since many of these variables are based on irrational factors such as human sentiment. However, I developed a model that uses a combination of sentiment analysis, clustering and feed-forward neural network in order to predict the market trends.