/
about.qmd
43 lines (33 loc) · 2.29 KB
/
about.qmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
---
title: "About"
image: https://avatars.githubusercontent.com/u/21018714?s=400&u=b47cb5eb401714a0e38aa8d32ec7031e5c30e346&v=4
about:
template: solana
image-shape: round
links:
- icon: twitter
text: Twitter
href: https://twitter.com/aniketmaurya
- icon: linkedin
text: LinkedIn
href: https://linkedin.com/in/aniketmaurya
- icon: github
text: Github
href: https://github.com/aniketmaurya
- icon: youtube
text: YouTube
href: https://youtube.com/@aiwithaniket
---
Hi there! I'm Aniket, a **Machine Learning - Software Engineer** with with over 4 years of experience, demonstrating a strong track record in developing and deploying machine learning models to production.
Currently, I'm working on building LLM-powered products and contributing to open source at Lightning AI ⚡️.
At Lightning AI, I have been working on on a wide range of AI products and use-cases as well as lead the open-source community.
Some of my contributions include:
* Designed and developed the flagship image generation product using Stable Diffusion, [Muse](https://lightning.ai/pages/community/tutorial/deploy-diffusion-models/).
* Increased model serving throughput efficiency by more than *50%* by implementing dynamic batching.
* Designed and developed *[Chat with PDF](https://lightning.ai/lightning-ai/studios/document-chat-assistant-using-rag)*, a RAG application using Large Language Models (LLMs).
* Enabled LLM evaluation by integrating Eval-Harness framework and Stanford's HELM with [Lit-GPT](https://github.com/Lightning-AI/lit-gpt).
Before joining Lightning AI, I was at [Quinbay](https://www.linkedin.com/company/quinbay/), a startup based in Bangalore, where I built and deployed deep learning models in production for our partner Blibli.com.
* Built and deployed Image Recognition system to block objectionable content, scaled for 10M requests.
* Reduced the Merchant Onboarding time from 24 hours to 1 second with a real-time ID Card Verification & Autofill System.
* Introduced a Face Recognition System for fraud detection in the e-commerce wallet, which reduced the manual process by more than 50%.
You can find most of my new blogs on LLMs, Distributed Training, and Machine Learning on [Lightning AI's blog](https://lightning.ai/pages/author/aniket-maurya/).