Artificial Intelligence

2023-12-13 09:55:54 Microsoft releases Phi-2, a small language model AI that outperforms Llama 2, Mistral 7B

The rapid pace of generative AI news and announcements isn’t slowing down, even as we reach the final stretches of 2023 and the traditional winter holiday quiet period.

Just take a look at Microsoft Research, the blue sky division of the software giant, which today announced the release of its Phi-2 small language model (SML), a text-to-text AI program that is “small enough to run on a laptop or mobile device."
https://www.microsoft.com/en-us/research/blog/phi-2-the-surprising-power-of-small-language-models/ Open / Comment

2023-12-11 17:43:01 AI Engineers vs Software Engineers:

Software engineers are often shocked when they learn of AI engineers' salaries. There are two reasons for this surprise.

1. The total compensation for AI engineers is jaw-dropping. You can check it out at AIPaygrad.es, which has manually verified data for AI engineers. The median overall compensation for a “Novice” is $328,350/year.
2. AI engineers are no smarter than software engineers. You figure this out only after a friend or acquaintance upskills and finds a lucrative AI job.

The biggest difference between Software and AI engineers is the demand for such roles. One role is declining, and the other is reaching stratospheric heights.

Here is an example.

Just last week, we saw an implosion of OpenAI after Sam Altman was unceremoniously removed from his CEO position. About 95% of their AI Engineers threatened to quit in protest. Rumor had it that these 700 engineers had an open job offer from Microsoft.

Contrast this with the events a few months back. Microsoft laid off 10,000 Software Engineers while setting aside $10B to invest in OpenAI. They cut these jobs despite making stunning profits in 2023.

In conclusion, these events underline a significant shift in the tech industry. For software engineers, it's a call to adapt and possibly upskill in AI, while companies need to balance AI investments with nurturing their current talent. The future of tech hinges on flexibility and continuous learning for everyone involved."

Join the forefront of the AI revolution with our AI Bootcamp. Learn from experts who have worked at big companies like AWS, Facebook, and Deloitte and studied at top colleges like IITs, BITS Pilani & COEP. The course lasts 26 weeks and includes 9 different projects and over 15 tools. Classes are online on weekends. Great deal for students at only ₹9999.
https://forms.gle/9DxhEgmcNEt29orH8 Open / Comment

2023-12-09 11:10:54 Artificial Intelligence Bootcamp by IIT & COEP Alumni

+90 hrs sessions.
+26 Weeks.
+15 Tools & Technology.
+9 Case studies.
+9 Projects.
+9 Skills.
+7 Domains.

Duration: 6 months

Remote and weekend Sessions.
Starting from basics.
Get Certificate / Internship Letter.

For registrations: https://forms.gle/9DxhEgmcNEt29orH8 Open / Comment

2023-12-07 06:10:04 Google finally did it!

It shipped a model that beats OpenAI's GPT-4

Here's all you need to know about Gemini & why it will change the AI game:

Gemini is Google's most capable and general AI model yet. It's a family of multimodal models that demonstrate really strong capabilities across the image, audio, video, and text domains.

In other words, it can easily recognize images, speak in real-time, has an incredible understanding of science, demonstrates sophisticated reasoning and advanced coding. Checkout

Gemini comes in three sizes:
1. Ultra for complex tasks
2. Pro for scaling across a range of tasks
3 Nano for efficient on-device tasks.

Now here comes the crazy part.

Gemini Ultra’s performance beats current state-of-the-art results in 30 of 32 benchmarks used in LLM research and development. More importantly, it's the first model to beat OpenAI's GPT-4, the dominant force in AI today

The age of AI has never been more exciting. Open / Comment

2023-12-04 16:17:01 𝐎𝐩𝐞𝐧-𝐬𝐨𝐮𝐫𝐜𝐞 𝐋𝐋𝐌 is free & available for anyone to access, & developers and researchers are free to use, improve, or modify the model.

𝐏𝐫𝐨𝐩𝐫𝐢𝐞𝐭𝐚𝐫𝐲 𝐋𝐋𝐌 is owned by a company & may include a license that restricts how the model can be used.

Generally, proprietary LLMs have far larger parameter sizes than open-source models.

But this is not always the case, Falcon 180B, the largest open-access model, with 180B parameters, surpassing most known LLMs.

The size of parameters in an LLM is a measure of its complexity.

In general, LLMs with more parameters tend to perform better on a variety of tasks.

Open-source LLMs have their own ecosystem, with unique advantages.

Open-source LLMs offer 𝐠𝐫𝐞𝐚𝐭𝐞𝐫 𝐭𝐫𝐚𝐧𝐬𝐩𝐚𝐫𝐞𝐧𝐜𝐲, allowing us to gain better insights into the model's workings, architecture & training data used.

Another advantage of open-source LLMs is the ability to benefit from 𝐜𝐨𝐦𝐦𝐮𝐧𝐢𝐭𝐲 𝐜𝐨𝐧𝐭𝐫𝐢𝐛𝐮𝐭𝐢𝐨𝐧𝐬. Researchers & developers can experiment and utilize contributions from individuals with diverse perspectives. Open / Comment

2023-12-01 16:12:01 𝗖𝘂𝘀𝘁𝗼𝗺 𝗚𝗣𝗧 𝗶𝘀 𝗴𝗼𝗻𝗻𝗮 𝗯𝗲 𝗮 𝗠𝘂𝗹𝘁𝗶-𝗠𝗶𝗹𝗹𝗶𝗼𝗻 𝗢𝗽𝗽𝗼𝗿𝘁𝘂𝗻𝗶𝘁𝘆.

Custom GPTs are starting to take off and become incredibly useful.

I believe the GPT Store will create multi-millionaires like the Apple app store did. Here are the top Custom GPTs I've found so far:

1. Grimoire
2. ResearchGPT
3. Storybook Vision
4. VideoGPT
5. AIProductGPT
6. The Secret of Monkey Island: Amsterdam
7. DALL-E 3 With Parameter
8. Custom Logo
9. WrappedGPT Open / Comment

2023-11-28 17:49:57 This Harvard Business School study found Boston Consulting Group consultants who used GPT-4 in their work finished on average 12% more tasks, 25% faster, with 40% higher quality. This is the potential of AI. https://www.thecrimson.com/article/2023/10/13/jagged-edge-ai-bcg/ Open / Comment

2023-11-26 13:39:48 This hits hard.

This AI security/privacy awareness video is a must watch.

Imagine the impact on your loved ones.

AI has incredible potential, but the risks are real.

This is why we need serious conversation around AI Safety and education. Open / Comment

2023-11-22 16:03:28 AI Engineers vs Software Engineers:

Software engineers are often shocked when they learn of AI engineers' salaries. There are two reasons for this surprise.

1. The total compensation for AI engineers is jaw-dropping. You can check it out at AIPaygrad.es, which has manually verified data for AI engineers. The median overall compensation for a “Novice” is $328,350/year.
2. AI engineers are no smarter than software engineers. You figure this out only after a friend or acquaintance upskills and finds a lucrative AI job.

The biggest difference between Software and AI engineers is the demand for such roles. One role is declining, and the other is reaching stratospheric heights.

Here is an example.

Just last week, we saw an implosion of OpenAI after Sam Altman was unceremoniously removed from his CEO position. About 95% of their AI Engineers threatened to quit in protest. Rumor had it that these 700 engineers had an open job offer from Microsoft.

Contrast this with the events a few months back. Microsoft laid off 10,000 Software Engineers while setting aside $10B to invest in OpenAI. They cut these jobs despite making stunning profits in 2023.

In conclusion, these events underline a significant shift in the tech industry. For software engineers, it's a call to adapt and possibly upskill in AI, while companies need to balance AI investments with nurturing their current talent. The future of tech hinges on flexibility and continuous learning for everyone involved." Open / Comment

2023-11-21 06:31:07 10 Things you need to become an AI/ML engineer:

1. Framing machine learning problems
2. Weak supervision and active learning
3. Processing, training, deploying, inference pipelines
4. Offline evaluation and testing in production
5. Performing error analysis. Where to work next
6. Distributed training. Data and model parallelism
7. Pruning, quantization, and knowledge distillation
8. Serving predictions. Online and batch inference
9. Monitoring models and data distribution shifts
10. Automatic retraining and evaluation of models

Join our WhatsApp Channel. Open / Comment