---
title: "Harshit Sharma (@sparkai) | Hashnode"
description: "Harshit Sharma (@sparkai) writes on Hashnode. Read their latest articles and follow their work."
url: "https://hashnode.com/@sparkai"
---

@sparkai

# Harshit Sharma

@sparkai · Joined March 2025

### [Turbo Quant : Visual Essay Series](https://connectai.blog/turbo-quant-visual-essay-series?utm_source=hashnode&utm_medium=feed)

https://turboquant-blogs.vercel.app/

1 min read · [connectai.blog](https://connectai.blog/turbo-quant-visual-essay-series?utm_source=hashnode&utm_medium=feed)

[Apr 14](https://hashnode.com/posts/turbo-quant-visual-essay-series/69de7dfa91716f3cfb63f38f)

### [Agents : Architecture and Design](https://connectai.blog/agents-architecture-and-design?utm_source=hashnode&utm_medium=feed)

Ai agents : Architecture and Design

2 min read · [connectai.blog](https://connectai.blog/agents-architecture-and-design?utm_source=hashnode&utm_medium=feed)

[Mar 15, 2025](https://hashnode.com/posts/agents-architecture-and-design/67d5329a2e244e73d1efae4c)

### [Agents : Memory](https://connectai.blog/agents-memory?utm_source=hashnode&utm_medium=feed)

Implement memory in Agents

13 min read · [connectai.blog](https://connectai.blog/agents-memory?utm_source=hashnode&utm_medium=feed)

[Mar 10, 2025](https://hashnode.com/posts/agents-memory/67ce80323296859ba874d9f6)

### [DeepGEMM: Clean and Efficient FP8 GEMM Library](https://connectai.blog/deepgemm-clean-and-efficient-fp8-gemm-library?utm_source=hashnode&utm_medium=feed)

Introduction DeepGEMM is a clean and efficient FP8 General Matrix Multiplication (GEMM) library with fine-grained scaling, released by DeepSeek as part of their "Open Source Week" in February 2025. It supports both normal dense GEMMs and Mixture-of-E...

4 min read · [connectai.blog](https://connectai.blog/deepgemm-clean-and-efficient-fp8-gemm-library?utm_source=hashnode&utm_medium=feed)

[Mar 8, 2025](https://hashnode.com/posts/deepgemm-clean-and-efficient-fp8-gemm-library/67cc844b47f860870a9950ba)

### [Understanding Load Balancing and Expert Parallelism in AI Models](https://connectai.blog/understanding-load-balancing-and-expert-parallelism-in-ai-models?utm_source=hashnode&utm_medium=feed)

Expert Parallelism Load Balancer (EPLB) In the world of AI and deep learning, managing the computational load efficiently is a critical task. Models, especially large-scale ones like those used in Natural Language Processing (NLP) or computer vision,...

5 min read · [connectai.blog](https://connectai.blog/understanding-load-balancing-and-expert-parallelism-in-ai-models?utm_source=hashnode&utm_medium=feed)

[Mar 5, 2025](https://hashnode.com/posts/understanding-load-balancing-and-expert-parallelism-in-ai-models/67c8a6f7cf2def7e4d146f9c)
