LLM fine-tuning techniques I'd learn if I were to customize them:
Bookmark this.
1. LoRA
2. QLoRA
3. Prefix Tuning
4. Adapter Tuning
5. Instruction Tuning
6. P-Tuning
7. BitFit
8. Soft Prompts
9. RLHF
10. RLAIF
11. DPO (Direct Preference Optimization)
12. GRPO (Group Relative Policy Optimization)
13. RLAIF (RL with AI Feedback)
14. Multi-Task Fine-Tuning
15. Federated Fine-Tuning
Since we're talking about fine-tuning, I wrote a full breakdown on fine-tuning LLMs with RL in 2026. Including how to skip manual reward engineering with automatic LLM-graded rewards.
And this is done using a 100% open-source solution: https://t.co/w1KJD7LZWe
The article is quoted below.
Google just dropped a 1-hour course on agentic engineering from scratch:
00:00 โ How to build your first AI agent
08:24 โ Build agent memory (short, persistent, long)
28:34 โ Agentic loops, long-running AI agents
40:04 โ How to build MCP (MCP vs API)
1:00:22 โ Multi-agentic systems
This 1-hour watch will replace 10 paid agentic courses on the internet.
Watch it today, then read how to build a self-improving agentic system in the article below.
Performance Hints
Over the years, my colleague Sanjay Ghemawat and I have done a fair bit of diving into performance tuning of various pieces of code. We wrote an internal Performance Hints document a couple of years ago as a way of identifying some general principles and we've recently published a version of it externally.
We'd love any feedback you might have!
Read the full doc at: https://t.co/jej95g236P
@airtelindia absolutely disappointed with auto-renewal of AmazonPrime subscription without an unsubscribe option in AirtelThanks app. Even more frustrating is the response from customer care suggesting waiting for deduction in the next bill cycle for refund request. Unacceptable!
Chandrayaan-3 Mission:
'India๐ฎ๐ณ,
I reached my destination
and you too!'
: Chandrayaan-3
Chandrayaan-3 has successfully
soft-landed on the moon ๐!.
Congratulations, India๐ฎ๐ณ!
#Chandrayaan_3#Ch3
Hey @Apple, please remove the excessive post processing on the iPhone 14 camera! It's making the sharpness too exposed and unnatural. Disappointed with camera experience. #iPhone14#CameraIssues#PostProcessing
Just Released
Deepest-ever view of the cosmos from @NASAWebb. The spiked objects are local stars in our own Galaxy. ignore them. Everything else is an entire galaxy. Many distort into arcs, revealing spacetime curvature from the gravity of a galaxy cluster in the image's center.
@MicrosoftTeams Kudos for introducing the chat with yourself for drafts, sending files, etc. An extremely beneficial feature.
@WhatsApp waiting for the addition of this feature in the app too.
@DGCAIndia@MoCA_GoI Booked 4 tickets from New Delhi to Vienna with @goibibo and @Aeroflot_World, with travel dates on March 29, 2020 PNR: S48GIV. Due to nationwide lockdown we received the credit shell from @Aeroflot_World airlines for 3 years. (1/2)
However, since the airline is still non-operational we are unable to redeem the credit shells with expiry approaching. Request your kind assistance in getting the refund as our requests for refund are getting rejected by airlines. (2/2)
@WhatsApp I think there is a need to have a "Notes to self" feature in the application. The current way of finding one's own number from contacts and sending messages/notes is inconsistent, especially in iPhone.
To find your answers, you need to ask the right questions. A good question is an *actionable question*: one that will actively guide you towards useful ideas and that will expand the set of solutions you consider.
People who say "Machine Learning is easy" are way more famous on Twitter.๐คฆ๐คฆ
Don't know when I will get that notion but today, even after 4 years working in the field of ML, I still find it harder than a lot of other things that I do!
Mental health experts urge media to exercise restraint while reporting #Covid-19 without compromising on facts or public interest. I would urge every media person to read this.