Vishal N S

Senior Consultant | Leading AI Solutions and Strategic Innovation @Deloitte

[Vishal N S @ivishalns]

https://t.co/RsnMfWMPgG

[Justin Liao @imjustinliao]

whoever made this for design engineers, I really appreciate your service.

-> https://t.co/wecuVxKXWW

[Kasturi @itskasturiii]

If you're out of ideas for loaders, I've open sourced text streaming, in-line loaders and image loaders. All free. Have fun.

https://t.co/O5ynnq6q2y
npm install generative-loaders

[Adi @AdityaSur11]

I made an Age Progression slider.

Not perfect i ran out of patience of refining the edges lol.

[rari @0xwhrrari]

Google just dropped a free 1-hour course

On building complete agent systems from scratch:

07:02 - Build your first AI agent
13:34 - Give every agent a specialized skill
23:08 - Engineer the right context
32:56 - Connect agents into graphs
54:11 - Run 1,000 agents in autonomous loops

One agent saves you an hour

A thousand agents replace the team you cannot afford to hire

Most people are still learning how to prompt one chatbot

Google is already teaching how to build entire agent systems

Agent → Skills → Context → Graphs → Loops

Single agents are the old workflow

Agent swarms are the new one

This free 1-hour course is worth more than most paid agent engineering programs

Bookmark it before everyone starts building the same thing

Then build your first agent with the guide below

[Kamran Ahmed @nilbuild]

I just released 70K+ high-quality hand-drawn icons, completely free to use.

Use them however you want. No attribution required.

Check them out 👇

[Roan @RohOnChain]

https://t.co/8HHG7saiut

[Codez @0xCodez]

Andrew Ng just released a 1-hour course on building agentic knowledge Graphs from scratch:

• 00:00 - Introduction to agentic knowledge Graphs
• 03:07 - Construction of agentic Graphs
• 14:00 - Architecture of multi-agent systems
• 23:00 - Building agentic graphs with Google ADK
• 01:06:03 - Why Graphsare the future of agentic AI

Worth more than 10 articles on loop engineering.

Watch it today, then read how to become a graph engineer in the article below.

[Dhairya @dkare1009]

Google just dropped a 1-hour course on agentic engineering from scratch:

00:00 – How to build your first AI agent
08:24 – Build agent memory (short, persistent, long)
28:34 – Agentic loops, long-running AI agents
40:04 – How to build MCP (MCP vs API)
1:00:22 – Multi-agentic systems

This 1-hour watch will replace 10 paid agentic courses on the internet.

Bookmark this. Watch this weekend.

[MIKE @mikenevermiss]

Claude Fable 5 is actually a beast for web design.

this guy just dropped a full guide on how to design and build interactive animated 3D websites with Claude Code + Fable 5

(complete prompts in the comments)

watch the video and read the full guidelines on the article below

[codila @0xCodila]

Anthropic and Andrew Ng built an agent that uses 90% fewer tokens from scratch:

they dropped the entire book of Frankenstein into a prompt - 108,000 tokens asked one question

the input dropped from 108,000 tokens to 11

here's how:

step 1 → put everything that never changes at the top - tools, then system, then docs

step 2 → mark where the static part ends. everything above it gets cached

step 3 → one stray space breaks it - and you pay full price again

step 4 → the cache dies in 5 min - every read resets the clock

step 5 → cached tokens don't count against your rate limits. free headroom

most people never touch this - it pays for itself on day one

watch & bookmark - this 1-hour brilliant course ↓

[harpreet @DataScienceHarp]

satellite imagery tells you a vineyard exists.

it doesn't tell you which rows are overgrown, where the canopy is thinning, or how the terrain drains between vines

VineLiDAR: high-density RGB-colored 3D LiDAR point clouds from drone flights over commercial vineyards in northern Spain.

same two blocks captured across three seasons, two years, and three altitudes (20m, 30m, 50m). ten flights total

public UAV LiDAR datasets for agriculture barely exist. this is one of the first

parsed all ten flights into fiftyone as interactive 3D point clouds. filter by season, altitude, or vineyard block and watch the canopy structure change over time

check it out here: https://t.co/Z1IeEVEqSd

[Vikas gupta @vicky_grok]

Anthropic engineer:

"Don't just prompt Claude. Build a system that can prompt itself."

In this 45-minute session, she explains how Anthropic builds AI agents that can remember past work, learn from mistakes, and improve with every run.

It's one of the best free resources for learning how AI agents work.

Watch the session, then read How Browser Agents Automate Online Workflows

[Techniahqrobot | humanoid robots @techniahqrobot]

This is the Scorpion Hexapod, a six-legged robotic scorpion created at Ghent University UGent Campus Kortrijk in Belgium.

It was built by students Stephan Flamand Robbe Terryn and Pieterjan Deconinck as part of an Embedded Prototyping / Mechatronics Design project.

What it is

• A biomimetic robot inspired by the body and movement of a real scorpion
• A hexapod, meaning it walks on six legs
• A university prototype, not a commercial robot
• Built to test animal-inspired movement, sensors and interactive behavior
• Designed more for robotics research and education than real-world work

Main hardware

• 6 walking legs for crawling movement
• 2 front claws for the scorpion look
• Moving tail with a stinger-style mechanism
• Sensors in the body, legs and claws
• Arduino-based electronics for control
• Front camera and proximity sensing
• Battery pack for mobile operation
• 3D-printed modules for legs and tail
• Laser-cut ABS body parts
• Thermoformed shell for the white outer body

What it can do

• Walk across the floor using its six legs
• Move its tail like a real scorpion
• React when a person gets close
• Detect when someone covers its front sensors
• Strike with its tail in the demo
• Leave a red mark using a marker pen attached to the stinger
• Operate through remote control
• Perform some simple autonomous reactions

Why it was created

• To explore bio-inspired robotics
• To show how digital fabrication can produce complex moving robots
• To combine 3D printing, laser cutting, Arduino electronics and sensors
• To teach students how to design a complete mechatronic system
• To improve on an older robotic ant project that had weak autonomy, short battery life and motor problems
• To create an interactive robot that reacts to humans in a visible way

Important note

• It was not built for combat
• It was not made for industrial deployment
• It is not a military robot
• It is an educational robotics prototype made to demonstrate movement, sensing and interaction

[Khushi Agarwal @Aicoder786]

ANDREJ KARPATHY COULD HAVE CHARGED $2,000 FOR THIS COURSE.

He put it on YouTube.

The full training stack. Tokenization. Neural network internals. Hallucinations. Tool use. Reinforcement learning. RLHF. DeepSeek. AlphaGo.

3 hours of the most comprehensive LLM education that exists anywhere at any price.

Not how to use the tools.

How the entire system was built from the ground up and why it behaves the way it does.

The engineers who understand this build things the ones who only use the tools cannot even conceive of.

The gap between those two groups is not 3 hours.

It is everything those 3 hours quietly unlock for the rest of your career.

[aditya @adxtyahq]

“design a RAG pipeline for 10M docs with zero hallucination”

apparently this was asked in a Google L5 interview round. came across it somewhere on the internet and honestly it’s a way more interesting system design problem than most classic distributed systems questions

1. ingest + normalize docs
- remove duplicates, standardize formats, extract metadata, maintain version history

2. hybrid retrieval (BM25 + embeddings)
- BM25 handles exact keyword matching while embeddings capture semantic meaning
- semantic search alone usually struggles with precision at massive scale

3. ANN retrieval + reranking
- ANN (Approximate nearest neighbor ) quickly pulls top candidate chunks from millions of docs
- then a reranker rescoring step improves relevance by deeply comparing query vs retrieved chunks

4. source confidence scoring
- every retrieved chunk gets scored based on freshness, trust level, overlap and retrieval consistency
- low-confidence context should never heavily influence generation

5. constrained generation
- the model is only allowed to answer using retrieved context (nothing new to be invented outside of the retrieved context)

6. citation-backed responses
- every major claim links back to exact chunks, documents or timestamps

7. hallucination fallback layer
- if retrieval confidence drops below a threshold: “insufficient evidence found”

8. continuous evals
- run adversarial queries, retrieval recall benchmarks and hallucination tests continuously

9. caching + memory layer
- cache high-frequency enterprise queries and retrieval paths (improves latency and output)

10. observability everywhere
- trace retrieval paths, chunk rankings, token attribution and failure points

Also at 10M docs, retrieval quality matters more than the frontier model itself.

[zostaff @zostaff]

Jane Street, Goldman Sachs, JP Morgan, BlackRock, Hudson River Trading, Two Sigma, D.E. Shaw.

The most expensive engineering teams in the world released their financial tools on GitHub. Here are 7 repos, one from each.

1. Jane Street, janestreet/magic-trace
https://t.co/a2G20vnewK
5.3k stars. Process tracer powered by Intel PT. When your profiler is blind, magic-trace sees every CPU instruction.

2. Goldman Sachs, goldmansachs/gs-quant
https://t.co/SMYFwP3TWD
Derivative pricing the GS traders use at their desks. MIT licensed.

3. JP Morgan, finos/perspective
https://t.co/9rgy6FxYt4
What JPM traders use to watch markets in real time. A $24k/year terminal, for free.

4. BlackRock, blackrock/lcso
https://t.co/iHwsxZDZD9
Rust optimizer for portfolio problems. Where scipy gives up, this works.

5. Hudson River Trading, hudson-trading/corral
https://t.co/YhmrQFmYaZ
Structured concurrency for C++20. The foundation of HFT infrastructure at one of the largest U.S. trading firms.

6. Two Sigma, twosigma/flint https://t.co/ebEFqcDxJ6
Time-series joins on Apache Spark with temporal tolerance. Built for billions of ticks.

7. D.E. Shaw, deshaw/pyflyby https://t.co/uYDQKtnDVd
Auto-import for IPython and Jupyter. D.E. Shaw also funded the development of IPython itself.

Bookmarked it

[Viktor Oddy @viktoroddy]

❤️‍🔥Gemini 3.5 Flash is so cracked at animated websites.

Just Recorded a 25 min Tutorial on how to Design websites with Google AI Studio + Gemini 3.5 ($50k Website)

[Charly Wargnier @DataChaz]

🚨 STOP BURNING YOUR TOKENS!

If you use Claude Code, you are probably wasting 80% of your context window.

I found 10 ace tools that will completely rescue your API bill.

1. Caveman Claude
- Literally makes Claude talk like a caveman
- Slashes 75% of output tokens with zero loss in accuracy
Repo → https://t.co/eEvSOvHutG

2. RTK (Rust Token Killer)
- A blazing fast proxy that filters terminal output
- 60-90% reduction and completely dependency-free
Repo → https://t.co/lDfjbsbPD5

3. Code Review Graph
- Claude reads only what matters using a Tree-sitter graph
- An unbelievable 49x token reduction on huge monorepos
Repo → https://t.co/xGn6Pp88yX

4. Context Mode
- Sandboxes raw output into SQLite instead of your context
- A staggering 98% context reduction on logs & GitHub
Repo → https://t.co/Jut2bvBMUD

5. Claude Token Optimizer
- Brilliant setup prompts that optimize any project
- 90% token savings, taking docs from 11K to 1.3K
Repo → https://t.co/0uOFODbG7e

6. Token Optimizer
- Hunts down the invisible ghost tokens eating your context
- Fully restores and protects your context quality
Repo → https://t.co/LUOzjECXKm

7. Token Optimizer MCP
- Adds aggressive caching and compression to your MCP tools
- 95%+ token reduction through pure intelligence
Repo → https://t.co/b5Eqruo2PM

8. Claude Context
- Zilliz’s hybrid vector search MCP
- Makes your entire codebase the context for 40% less cost
Repo → https://t.co/hPG6pb0j3G

9. Claude Token Efficient
- Just drop one CLAUDE.md file into your repo
- Enforces strict terseness with zero code changes
Repo → https://t.co/fNrl6nwItF

10. Token Savior
- Navigates your code by symbols, not giant files
- 97% reduction on code navigation with persistent memory
Repo → https://t.co/lkILPhfwJh

----

[ The god-tier stack ]
Pick 2-3 based on what’s draining you:
> Massive repo? Code Review Graph + Token Savior
> Heavy terminal output? RTK
> MCP data dumps? Context Mode
> Need an instant fix? Caveman + Claude Token Efficient

Most devs are bleeding tokens.

Run `/context` in a fresh session and watch the savings roll in 👀

[Movez @0xMovez]

SNAPCHAT PAID $150,000,000 FOR LOOKSERY - STARTUP IN DEEP LEARNING COMPUTER VISION

This 1-hour Stanford lecture on "DL for Computer Vision" will teach you how to build same project from scratch.

Bookmark it & watch today. Stanford's full course (19 lectures) on Deep Learning for Computer Vision below ↓