kevalsaud

[kevalsaud @kevalsaud25]

I'm ready to bring my skills in Python and machine learning to a great team. If you're hiring for an entry-level or intern position, I'd love to connect!
#AI #MachineLearning #DataScience #Python #LLM #OpenToWork #Tech #GenerativeAI #Internship #Hiring #

[kevalsaud @kevalsaud25]

Seeking entry-level/intern roles in AI & ML!
I'm a CS Master's student passionate about building impactful products.

[kevalsaud @kevalsaud25]

3. Hands-on Projects: I led the development of "MedicLLM," an AI assistant for psychiatric diseases, which was tested by a renowned hospital . This project involved creating an argumentative multi-agent framework and XAI which improved clinician trust.

[kevalsaud @kevalsaud25]

2. I built a Part-of-Speech (POS) tagger for an under-resourced language. For this, I trained and compared a range of models, including classic SVM and HMM, as well as deep learning models. I also fine-tuned modern SLM deploying the final solution as a live Streamlit app.

[kevalsaud @kevalsaud25]

A few things I've worked on:

1. Professional Experience : At a previous organization, I developed a web mining application and fine-tuned Stable Diffusion models, which improved image generation. I also engineered RAG-based LLM services and built a conversational AI chatbot

[kevalsaud @kevalsaud25]

Was this the only thing left? Now even research papers are done by AI

In this new arXiv paper, “C. Opus” co-authors a rebuttal defending AI reasoning by blaming token limits and bad prompts.

Basically: AI just explained why humans were wrong about AI 😅

Self-defense: On

[kevalsaud @kevalsaud25]

Attended Techfest today and by IIT Bombay:
Udaan – enabling translation of learning materials across Indian languages with Project LeapOver bridging language barriers.
BharatGen – building Generative AI tailored to India’s needs with homegrown LLMs like BharatGen2B.

#techfest

[kevalsaud @kevalsaud25]

AutoDev By Microsoft and Software development agent.

https://t.co/QCw9IzEIRh

[kevalsaud @kevalsaud25]

In a mere 24 months, large language models have catalyzed an AI renaissance. Groundbreaking systems like LLMs.

Their ability to understand, analyze, and generate.This swift paradigm shift demonstrates AI's potential to reshape industries and redefine the boundaries of A.I

[kevalsaud @kevalsaud25]

It's here try it out ! @Google @GoogleDeepMind #GEMINIAI

[Balaji @balajis]

500+ OpenAI employees will quit and join Microsoft unless the board resigns and reinstates Sam and Greg.

[kevalsaud @kevalsaud25]

"Today, I had the pleasure of meeting a statue of the great Alan Turing. Even though he's cast in stone, his genius has a way of Turing people into computer enthusiasts like me! 🌟🤖 His legacy continues to inspire and shape the world of technology." #AlanTuring

[The Deep View @theDeepView]

We spend all week researching the top AI tools so you don't have to...

Get insights, tools, and news for free when you join our community of 80,000+ readers.

[kevalsaud @kevalsaud25]

Falcon 180B was trained on 3.5 trillion tokens on up to 4096 GPUs simultaneously, using Amazon SageMaker for a total of ~7,000,000 GPU hours. This means Falcon 180B is 2.5 times larger than Llama 2.

Stands between GPT 3.5 and 4.
Blog Post hugging face:- https://t.co/czTuakWhu3

[kevalsaud @kevalsaud25]

Many LLMs sprint ahead, pursuing general intelligence. Many strives to keep up the pace. Though the finish line is still far off, the leading LLMs run with optimism, hoping their progress will benefit humanity.

@OpenAI @AnthropicAI @MosaicML @huggingface

#LLMs #AI

[kevalsaud @kevalsaud25]

@matchaman11 @langchain Done

[Anil Chandra Naidu Matcha @matchaman11]

Calling all @langchain developers

There are no Langchain communities in Twitter

Since there is no community

I created one for developers

To interact and learn from each other

Plan to keep a small and focused LangChain group

More details below ↓

[kevalsaud @kevalsaud25]

The ability to reduce training parameters by x1000. With just a couple of datasets, it is now possible to achieve desirable outcomes by your own data to a pre-trained model. It's remarkable how a different computational approach can achieve impressive results with minimal effort

[Amazon India @amazonIN]

🎉🎉🎉