- I am currently working as a Web Solutions Engineer at Google gTech team.
- Previously I have worked as a Software Engineer at GeeksForGeeks.
- I maintain an Engineering Newsletter "Systems That Scale" which is subscribed by 30,000+ Engineers and Students. Check it Out!
- My Projects:
- 👓 HuggingFace BPE Tokenizer Visualizer: https://github.com/SauravP97/hf-tokenizer-visualizer
- Collection of notebooks for Fine-tuning LLMs from scratch: https://github.com/SauravP97/llm-finetune
- A decoder only Transformer implementing masked attention from scratch (Pytorch): https://github.com/SauravP97/toy-transformer
- I also have my YouTube channel where I post content related to tech.
- "Distributed Systems 101" series is live on my channel.
- "RAG and LangChain - Building Multi-agent workflows" playlist in live on my channel.
- "AI Engineering 101" playlist is live on my channel.
- Checkout my research:
- "Deep Researcher with Sequential Plan Reflection and Candidates Crossover (Deep Researcher Reflect Evolve)" here: https://arxiv.org/abs/2601.20843
- "A Hierarchical Tree-based approach for creating Configurable and Static Deep Research Agent (Static-DRA)" here: https://arxiv.org/abs/2512.03887
👨💻
Practicing the subtle art of not giving a bug!
Senior Web Solutions Engineer @ Google
Pinned Loading
-
Saurav-s-DSA-Templates
Saurav-s-DSA-Templates PublicMy Data Structures and Algorithms templates
-
Saurav-Low-Level-Design-Template
Saurav-Low-Level-Design-Template PublicLow Level Design template
-
-
tiny-stories-hf
tiny-stories-hf PublicTrain Tiny Stories dataset from the paper - "TinyStories: How Small Can Language Models Be and Still Speak Coherent English?"
-
Transformer implementation in Pytorc...
Transformer implementation in Pytorch from scratch 1"""2A decoder-only Transformer implementation in Pytorch from scratch.3The code is inspired from @karpathy's "nanoGPT" implementation.45The code is organized as follows:
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.


