Kannav Sethi
I'm an AI applications developer. I build systems that use large language models and write about the math behind them -- attention mechanisms, positional embeddings, and how floating point numbers break in interesting ways.
Currently at Locate Alpha. Previously at Seneca Polytechnic and Project Human City.
writing
- Controlling Probabilistic Outcomes by Ensuring a Deterministic SystemJul 2026
- Local Models, Unified Memory, and Why Quantization MattersMay 2026
- A (not so big) Primer on GPU InternalsMar 2026
- Floats and Precision in LLM TrainingFeb 2026
- Normalization TechniquesDec 2025
- Rotary Positional Embeddings and its MathDec 2025
- KV Cache, MQA, GQA and Attention!!!!Nov 2025
work
- Software Engineer, Locate Alphacurrent
- AI Engineer, Acclaim Abilitycurrent
- Product Builder, Seneca Polytechnicpreviously
- Software Dev Intern, Project Human Citypreviously
projects
- DialectMorph — a TypeScript transpiler for cross-language code conversion
- Llama 2 Fine-tune — efficient LLM fine-tuning using LoRA