me (@tmaralerdene) 's Twitter Profile
me

@tmaralerdene

HODOPHILE, SOFTWARE ENGINEER

ID: 100230608

calendar_today29-12-2009 13:55:44

979 Tweet

271 Takipçi

1,1K Takip Edilen

Hiten Shah (@hnshah) 's Twitter Profile Photo

You open an app. It asks for your email before you’ve seen anything. Then a permissions request. Then a loading spinner. By the fourth screen, you’re out. It’s easy to blame churn on onboarding friction. But what actually failed was the contract. Every interaction is a trade.

Traycer (@traycerai) 's Twitter Profile Photo

Cursor is powerful, but inefficient, tons of re-prompts and wasted requests. We fixed that by adding a planning layer to Cursor. It’s free and makes your requests 5x more efficient. Try it 👇

TuringPost (@theturingpost) 's Twitter Profile Photo

6 Essential reads on core AI/ML topics: ▪️ Foundations of LLMs by Tong Xiao and Jingbo Zhu ▪️ LLMs Post-Training: Surveying Techniques from Alignment to Reasoning ▪️ Agentic LLMs, a survey ▪️ A Survey of Context Engineering for LLMs ▪️ A Survey of Generative Categories and

6 Essential reads on core AI/ML topics:

▪️ Foundations of LLMs by Tong Xiao and Jingbo Zhu
▪️ LLMs Post-Training: Surveying Techniques from Alignment to Reasoning
▪️ Agentic LLMs, a survey
▪️ A Survey of Context Engineering for LLMs
▪️ A Survey of Generative Categories and
Javarevisited (@javarevisited) 's Twitter Profile Photo

5 Must-Read LLM Engineering Books in 2025 1. LLM Engineering Handbook - buff.ly/wogklbo 2. Building LLMs for Production - buff.ly/wjpOeTB 3. Build a Large Language Mode - buff.ly/DHp4ZR1 4. Hands-On Large Language Models -

5 Must-Read LLM Engineering Books in 2025          
1. LLM Engineering Handbook - buff.ly/wogklbo 
2. Building LLMs for Production - buff.ly/wjpOeTB
3. Build a Large Language Mode - buff.ly/DHp4ZR1
4. Hands-On Large Language Models -
Zach Morris Wilson (@eczachly) 's Twitter Profile Photo

Stop building streaming pipelines when your stakeholders request “real time” data! When they ask for realtime, always ask for acceptable latency: If the acceptable latency is >=1 hour, please just use batch! If the acceptable latency is between 10 minutes and 1 hour, use

Stop building streaming pipelines when your stakeholders request “real time” data!

When they ask for realtime, always ask for acceptable latency: 

If the acceptable latency is >=1 hour, please just use batch!

If the acceptable latency is between 10 minutes and 1 hour, use
Gabbar (@gabbbarsingh) 's Twitter Profile Photo

Always thought the way Calculus is thought in School is very flawed. Alienates a lot of people. This current generation is lucky. Watch:

Dr Milan Milanović (@milan_milanovic) 's Twitter Profile Photo

After over 20 years in software engineering, I've seen teams waste months building authentication, user management, and billing systems. The result? Less time building features that generate revenue. This is precisely what @hey_kinde solves. 🧵 Here is why I'm recommending

After over 20 years in software engineering, I've seen teams waste months building authentication, user management, and billing systems.

The result? Less time building features that generate revenue.

This is precisely what @hey_kinde solves.

🧵 Here is why I'm recommending
Aurimas Griciūnas (@aurimas_gr) 's Twitter Profile Photo

𝗠𝗖𝗣 plus 𝗔𝟮𝗔, here is how they complement each other 👇 Protocol wars continue to rage, let's understand how Googles A2A (Agent2Agent) protocol is different from MCP and how they complement each other (read till the end). 𝘔𝘰𝘷𝘪𝘯𝘨 𝘱𝘪𝘦𝘤𝘦𝘴 𝘪𝘯 𝘔𝘊𝘗: 𝟭. MCP

𝗠𝗖𝗣 plus 𝗔𝟮𝗔, here is how they complement each other 👇

Protocol wars continue to rage, let's understand how Googles A2A (Agent2Agent) protocol is different from MCP and how they complement each other (read till the end).

𝘔𝘰𝘷𝘪𝘯𝘨 𝘱𝘪𝘦𝘤𝘦𝘴 𝘪𝘯 𝘔𝘊𝘗:

𝟭. MCP
tulga.eth (@itulga) 's Twitter Profile Photo

Англиар нэг үг зөв дуудаж чадахгүй, гадаадад очоод гадны хүнтэй ажлаа ярьчихаж чадахгүй байж дэлгүүрийн брэнд сайн мэднэ гэж. Буудалд шөнөжин архидаад, өдөржин архи үнэртүүлж хурал дээр унтаад, орой нь дэлгүүр хэсч, казиноддог нийтлэг маршруттай.

Ruben Hassid (@rubenhssd) 's Twitter Profile Photo

NVIDIA just dropped paper exposing a $57 billion AI industry mistake. While Big Tech keeps pushing expensive LLMs like ChatGPT & Claude... Small language models handle 70% of AI agent work at 1/30th the cost. Here's why this changes everything: (hint: less is more)

NVIDIA just dropped paper exposing a $57 billion AI industry mistake.

While Big Tech keeps pushing expensive LLMs like ChatGPT & Claude...

Small language models handle 70% of AI agent work at 1/30th the cost.

Here's why this changes everything:

(hint: less is more)
Avi Chawla (@_avichawla) 's Twitter Profile Photo

Build human-like memory for your Agents (open-source)! Every agentic and RAG system struggles with real-time knowledge updates and fast data retrieval. Zep solves these issues with its continuously evolving and temporally-aware Knowledge Graph. Like humans, Zep organizes an

Lukasz Olejnik (@lukolejnik) 's Twitter Profile Photo

Chinese scientists have developed the best shortest-path algorithm in 41 years! A team from Tsinghua University has broken Dijkstra’s “sorting barrier” — the first improvement since 1984. The new algorithm runs in O(m \log^{2/3} n) time. Potential applications? Faster shorter

Chinese scientists have developed the best shortest-path algorithm in 41 years! A team from Tsinghua University has broken Dijkstra’s “sorting barrier” — the first improvement since 1984. 
The new algorithm runs in O(m \log^{2/3} n) time. Potential applications? Faster shorter
Skilled Coder (@theskilledcoder) 's Twitter Profile Photo

If it’s real-time → WebSockets If it’s scale → Kafka If it’s simplicity → REST If it’s chaos → GraphQL If it’s AI → Python If it’s infra → Go If it’s logs → ElasticSearch If it’s low-latency → Redis If it’s high-availability → Postgres If it’s streaming → Flink

Bilbing (@bilzbond16485) 's Twitter Profile Photo

Furqan Rydhan Demo: Prompt → Scent We built an AI scent printer from scratch in 4 days. You type a mood/song etc, our model composes a blend + timeline, the robot doses oils, and the diffuser runs. Would love to demo at Founders Inc Furqan Rydhan

Alex Prompter (@alex_prompter) 's Twitter Profile Photo

Holy shit. MIT just built an AI that can rewrite its own code to get smarter 🤯 It’s called SEAL (Self-Adapting Language Models). Instead of humans fine-tuning it, SEAL reads new info, rewrites it in its own words, and runs gradient updates on itself literally performing

Holy shit. MIT just built an AI that can rewrite its own code to get smarter  🤯

It’s called SEAL (Self-Adapting Language Models).

Instead of humans fine-tuning it, SEAL reads new info, rewrites it in its own words, and runs gradient updates on itself literally performing
Bilbing (@bilzbond16485) 's Twitter Profile Photo

Hey #SFcouples! 🎶 Ready for the most unique date night? ✨ Join our Bottle Your Song: A Couple's AI Perfumery Workshop at Fort Mason! Translate your song into a custom perfume for each of you, guided by a chemist & AI. Perfect before/after the Sunday Farmers Market! 🗓️ Oct 19.