Jakub Slys
@iam.slys.dev
𝑇ℎ𝑒 𝑏𝑒𝑠𝑡 𝑡𝑖𝑚𝑒 𝑡𝑜 𝑝𝑙𝑎𝑛𝑡 𝑎 𝑡𝑟𝑒𝑒 𝑤𝑎𝑠 20 𝑦𝑒𝑎𝑟𝑠 𝑎𝑔𝑜. 𝑻𝒉𝒆 𝒔𝒆𝒄𝒐𝒏𝒅 𝒃𝒆𝒔𝒕 𝒕𝒊𝒎𝒆 𝒊𝒔 𝒏𝒐𝒘.
Sharding is like opening new branches when one library gets too busy. Users get speed, architects get headaches, and everyone gets a lesson in distributed systems. Ever sharded and regretted it? Share your story. Read more: https://iam.slys.dev/p/why-databases-get-split-sharding
Why databases get split — sharding, partitioning, and replication without fear
System design fundamentals
iam.slys.dev
February 3, 2026 at 8:21 PM
Sharding is like opening new branches when one library gets too busy. Users get speed, architects get headaches, and everyone gets a lesson in distributed systems. Ever sharded and regretted it? Share your story. Read more: https://iam.slys.dev/p/why-databases-get-split-sharding
Physics and AI, usually at odds, find a new harmony in particle-guided diffusion models. It’s a subtle shift: solutions that honor both creativity and physical laws. Could this approach unlock new possibilities for digital experimentation?
AI physics diffusionmodels scientificcomputing
AI physics diffusionmodels scientificcomputing
Particle-Guided Diffusion Models for Partial Differential Equations
arxiv.org
February 3, 2026 at 3:16 PM
Physics and AI, usually at odds, find a new harmony in particle-guided diffusion models. It’s a subtle shift: solutions that honor both creativity and physical laws. Could this approach unlock new possibilities for digital experimentation?
AI physics diffusionmodels scientificcomputing
AI physics diffusionmodels scientificcomputing
Ever wondered what keeps massive codebases like Chromium and LLVM robust? It’s frameworks like GoogleTest: auto test discovery, flexible assertions, and parameterized tests. Makes me rethink how much trust I put in my current unit tests.
Cplusplus testing infrastructure quality
Cplusplus testing infrastructure quality
GitHub - google/googletest: GoogleTest - Google Testing and Mocking Framework
GoogleTest - Google Testing and Mocking Framework. Contribute to google/googletest development by creating an account on GitHub.
github.com
February 3, 2026 at 12:13 PM
Ever wondered what keeps massive codebases like Chromium and LLVM robust? It’s frameworks like GoogleTest: auto test discovery, flexible assertions, and parameterized tests. Makes me rethink how much trust I put in my current unit tests.
Cplusplus testing infrastructure quality
Cplusplus testing infrastructure quality
Pretrained forecasting models are now versatile: a single embedding unlocks forecasting, classification, and anomaly detection across domains. Is true cross-domain modeling within reach? Read more: https://iam.slys.dev/p/real-world-agent-benchmarks-interpretable
✨ Real-world agent benchmarks, interpretable TSP solvers, and time-series foundation models
AI Alert
iam.slys.dev
February 2, 2026 at 8:21 PM
Pretrained forecasting models are now versatile: a single embedding unlocks forecasting, classification, and anomaly detection across domains. Is true cross-domain modeling within reach? Read more: https://iam.slys.dev/p/real-world-agent-benchmarks-interpretable
What if neural nets had more than one “winning ticket”—each one tuned to a different domain of reality? RTL recasts network pruning as a tool for building modular, context-sensitive models. Could this spark a shift toward truly adaptive AI?
AI networks datadiversity adaptation
AI networks datadiversity adaptation
Routing the Lottery: Adaptive Subnetworks for Heterogeneous Data
arxiv.org
February 2, 2026 at 3:16 PM
What if neural nets had more than one “winning ticket”—each one tuned to a different domain of reality? RTL recasts network pruning as a tool for building modular, context-sensitive models. Could this spark a shift toward truly adaptive AI?
AI networks datadiversity adaptation
AI networks datadiversity adaptation
There’s a new way to investigate and learn from websites—web-check. Built for anyone curious, it shows how much open data sits just beneath the surface. Makes me rethink how I approach site analysis!
websecurity devtools infosec opensource
websecurity devtools infosec opensource
GitHub - Lissy93/web-check: 🕵️♂️ All-in-one OSINT tool for analysing any website
🕵️♂️ All-in-one OSINT tool for analysing any website - Lissy93/web-check
github.com
February 2, 2026 at 12:14 PM
There’s a new way to investigate and learn from websites—web-check. Built for anyone curious, it shows how much open data sits just beneath the surface. Makes me rethink how I approach site analysis!
websecurity devtools infosec opensource
websecurity devtools infosec opensource
Python devs: ever puzzled why your multicore app isn’t scaling? The GIL might be the bottleneck forcing threads to take turns. How have you navigated its limits? Share your experience. Read more: https://iam.slys.dev/p/understanding-locking-contention
Understanding locking contention in computing
System Design
iam.slys.dev
February 1, 2026 at 8:21 PM
Python devs: ever puzzled why your multicore app isn’t scaling? The GIL might be the bottleneck forcing threads to take turns. How have you navigated its limits? Share your experience. Read more: https://iam.slys.dev/p/understanding-locking-contention
What do we miss when we only use the most downloaded models? Some of the best answers are lost in the noise, just waiting for more efficient ways to be found. Here’s to new methods that expand what’s possible in public AI.
AI opensource research modelrepositories
AI opensource research modelrepositories
Discovering Hidden Gems in Model Repositories
arxiv.org
February 1, 2026 at 3:16 PM
What do we miss when we only use the most downloaded models? Some of the best answers are lost in the noise, just waiting for more efficient ways to be found. Here’s to new methods that expand what’s possible in public AI.
AI opensource research modelrepositories
AI opensource research modelrepositories
It’s fascinating how Kiro uses structured specs and automation hooks to rethink the IDE. Natural language isn’t just for codegen—it’s changing how we collaborate and steer projects. Would you trust an AI-powered IDE with your dev flow?
automation productivity AI
automation productivity AI
GitHub - kirodotdev/Kiro: Kiro is an agentic IDE that works alongside you from prototype to production.
Kiro is an agentic IDE that works alongside you from prototype to production. - kirodotdev/Kiro
github.com
February 1, 2026 at 12:14 PM
It’s fascinating how Kiro uses structured specs and automation hooks to rethink the IDE. Natural language isn’t just for codegen—it’s changing how we collaborate and steer projects. Would you trust an AI-powered IDE with your dev flow?
automation productivity AI
automation productivity AI
It’s easy to trust the crowd when choosing from model repositories, but what if the crowd is missing out? Evidence now shows lesser-known models can outperform the top downloads. Smarter algorithms could help us uncover what we’re all overlooking.
AI ModelDiscovery HiddenGems
AI ModelDiscovery HiddenGems
Discovering Hidden Gems in Model Repositories
arxiv.org
January 31, 2026 at 10:46 PM
It’s easy to trust the crowd when choosing from model repositories, but what if the crowd is missing out? Evidence now shows lesser-known models can outperform the top downloads. Smarter algorithms could help us uncover what we’re all overlooking.
AI ModelDiscovery HiddenGems
AI ModelDiscovery HiddenGems
It’s easy to trust the crowd when choosing from model repositories, but what if the crowd is missing out? Evidence now shows lesser-known models can outperform the top downloads. Smarter algorithms could help us uncover what we’re all overlooking.
AI ModelDiscovery HiddenGems
AI ModelDiscovery HiddenGems
January 31, 2026 at 10:41 PM
It’s easy to trust the crowd when choosing from model repositories, but what if the crowd is missing out? Evidence now shows lesser-known models can outperform the top downloads. Smarter algorithms could help us uncover what we’re all overlooking.
AI ModelDiscovery HiddenGems
AI ModelDiscovery HiddenGems
Truth travels in pieces—distributed systems don’t always agree right away, and that's okay. The next time your data lags or changes seem out of sync, it's just the reality of modern networks catching up. Have you noticed this too?
Read more: https://iam.slys.dev/p/what-it-means-for-a-system-to-be
Read more: https://iam.slys.dev/p/what-it-means-for-a-system-to-be
What it means for a system to be consistent — and why it does not always have to be
System design fundamentals
iam.slys.dev
January 31, 2026 at 3:16 PM
Truth travels in pieces—distributed systems don’t always agree right away, and that's okay. The next time your data lags or changes seem out of sync, it's just the reality of modern networks catching up. Have you noticed this too?
Read more: https://iam.slys.dev/p/what-it-means-for-a-system-to-be
Read more: https://iam.slys.dev/p/what-it-means-for-a-system-to-be
Ever wondered if an AI could run your stock portfolio? This project hands that challenge to ChatGPT, tracking every gain, loss, and learning moment along the way. Trying to bridge hype with actual data. Do you trust an LLM with your trades?
AI investing openexperiment transparency
AI investing openexperiment transparency
GitHub - LuckyOne7777/LLM-Trading-Lab: This repo powers my experiment where ChatGPT manages a real-money micro-cap stock portfolio.
This repo powers my experiment where ChatGPT manages a real-money micro-cap stock portfolio. - LuckyOne7777/LLM-Trading-Lab
github.com
January 31, 2026 at 12:14 PM
Ever wondered if an AI could run your stock portfolio? This project hands that challenge to ChatGPT, tracking every gain, loss, and learning moment along the way. Trying to bridge hype with actual data. Do you trust an LLM with your trades?
AI investing openexperiment transparency
AI investing openexperiment transparency
Concurrent data structures are underrated upgrades. Why reinvent the wheel? Switching from a DIY lock to a battle-tested concurrent collection can make life easier. Ever swapped one out and never looked back? Read more: https://iam.slys.dev/p/understanding-locking-contention
Understanding locking contention in computing
System Design
iam.slys.dev
January 30, 2026 at 8:22 PM
Concurrent data structures are underrated upgrades. Why reinvent the wheel? Switching from a DIY lock to a battle-tested concurrent collection can make life easier. Ever swapped one out and never looked back? Read more: https://iam.slys.dev/p/understanding-locking-contention
Wild to see a 100B LLM humming along on a single CPU using BitNet. Is inferring at human reading speeds on commodity hardware the tipping point for truly local, personal AI? What tools do you want next?
LocalAI AIInfrastructure Innovation PowerEfficiency
LocalAI AIInfrastructure Innovation PowerEfficiency
GitHub - microsoft/BitNet: Official inference framework for 1-bit LLMs
Official inference framework for 1-bit LLMs. Contribute to microsoft/BitNet development by creating an account on GitHub.
github.com
January 30, 2026 at 5:39 PM
Wild to see a 100B LLM humming along on a single CPU using BitNet. Is inferring at human reading speeds on commodity hardware the tipping point for truly local, personal AI? What tools do you want next?
LocalAI AIInfrastructure Innovation PowerEfficiency
LocalAI AIInfrastructure Innovation PowerEfficiency
It fascinates me how PowerToys blends old-school utilities with AI, batch processing, and accessible design. The real magic is how the community shapes its evolution—OS-level tools have never felt more personal.
WindowsTools Community Productivity DeskLife
WindowsTools Community Productivity DeskLife
GitHub - microsoft/PowerToys: Microsoft PowerToys is a collection of utilities that help you customize Windows and streamline everyday tasks
Microsoft PowerToys is a collection of utilities that help you customize Windows and streamline everyday tasks - microsoft/PowerToys
github.com
January 30, 2026 at 5:32 PM
It fascinates me how PowerToys blends old-school utilities with AI, batch processing, and accessible design. The real magic is how the community shapes its evolution—OS-level tools have never felt more personal.
WindowsTools Community Productivity DeskLife
WindowsTools Community Productivity DeskLife
Ever wondered what keeps massive codebases like Chromium and LLVM robust? It’s frameworks like GoogleTest: auto test discovery, flexible assertions, and parameterized tests. Makes me rethink how much trust I put in my current unit tests.
Cplusplus testing infrastructure quality
Cplusplus testing infrastructure quality
GitHub - google/googletest: GoogleTest - Google Testing and Mocking Framework
GoogleTest - Google Testing and Mocking Framework. Contribute to google/googletest development by creating an account on GitHub.
github.com
January 26, 2026 at 12:13 PM
Ever wondered what keeps massive codebases like Chromium and LLVM robust? It’s frameworks like GoogleTest: auto test discovery, flexible assertions, and parameterized tests. Makes me rethink how much trust I put in my current unit tests.
Cplusplus testing infrastructure quality
Cplusplus testing infrastructure quality
This whole publication is basically my public growth log—sharing what’s worked and what’s hurt in the hope it’s useful to someone else. What’s the best lesson you’ve picked up here?
Get Your Cert
💰 Knowledge pays 💰 Literally.
iam.slys.dev
January 25, 2026 at 8:22 PM
This whole publication is basically my public growth log—sharing what’s worked and what’s hurt in the hope it’s useful to someone else. What’s the best lesson you’ve picked up here?
Failure isn’t a bug—it's the default in distributed systems. Real reliability means making chaos uneventful for the user. What surprises have you seen systems quietly absorb? Let’s swap stories. Read more: https://iam.slys.dev/p/how-systems-handle-failure-retries
How systems handle failure — retries, circuit breakers, and idempotency
System design fundamentals
iam.slys.dev
January 25, 2026 at 3:17 PM
Failure isn’t a bug—it's the default in distributed systems. Real reliability means making chaos uneventful for the user. What surprises have you seen systems quietly absorb? Let’s swap stories. Read more: https://iam.slys.dev/p/how-systems-handle-failure-retries
Is AI ready to be your code reviewer, project assistant, and trend spotter? Claude Code Action makes it as simple as mentioning @claude. I’m curious how this shifts trust in our dev routines.
LLMs DevOps Productivity AIforCode
LLMs DevOps Productivity AIforCode
GitHub - anthropics/claude-code-action
Contribute to anthropics/claude-code-action development by creating an account on GitHub.
github.com
January 25, 2026 at 12:13 PM
Is AI ready to be your code reviewer, project assistant, and trend spotter? Claude Code Action makes it as simple as mentioning @claude. I’m curious how this shifts trust in our dev routines.
LLMs DevOps Productivity AIforCode
LLMs DevOps Productivity AIforCode
Fetch latency can hinge on cache state machines. Intel’s MESIF protocol, with its special Forward state, decides who shares data. Ever architected for cache coherence? Let’s swap stories. More here: https://iam.slys.dev/p/cpu-caches-why-tiny-memory-matters
CPU Caches: Why tiny memory matters?
System design
iam.slys.dev
January 24, 2026 at 3:17 PM
Fetch latency can hinge on cache state machines. Intel’s MESIF protocol, with its special Forward state, decides who shares data. Ever architected for cache coherence? Let’s swap stories. More here: https://iam.slys.dev/p/cpu-caches-why-tiny-memory-matters
Observability is the secret ingredient behind every robust system. If you can’t see failure coming, you can’t react. How are you measuring system health? Read more: https://iam.slys.dev/p/how-systems-handle-failure-retries
How systems handle failure — retries, circuit breakers, and idempotency
System design fundamentals
iam.slys.dev
January 23, 2026 at 8:22 PM
Observability is the secret ingredient behind every robust system. If you can’t see failure coming, you can’t react. How are you measuring system health? Read more: https://iam.slys.dev/p/how-systems-handle-failure-retries
Speed is all about memory. Space-based systems keep data in-memory to slash latency for real-time apps. If you’ve ever worked with in-memory grids, you know how much happier users can get. Has in-memory changed your app experience?
Space-Based Architecture
Software architecture
iam.slys.dev
January 23, 2026 at 3:17 PM
Speed is all about memory. Space-based systems keep data in-memory to slash latency for real-time apps. If you’ve ever worked with in-memory grids, you know how much happier users can get. Has in-memory changed your app experience?
Ever wish you could just browse an index of every cool API out there? marcelscruz/public-apis is exactly that—a living, open list where new discoveries are just a scroll away. Which API exploration changed your coding direction?
APIs openweb programmers techcommunity
APIs openweb programmers techcommunity
GitHub - marcelscruz/public-apis: A collaborative list of public APIs for developers
A collaborative list of public APIs for developers - marcelscruz/public-apis
github.com
January 23, 2026 at 12:13 PM
Ever wish you could just browse an index of every cool API out there? marcelscruz/public-apis is exactly that—a living, open list where new discoveries are just a scroll away. Which API exploration changed your coding direction?
APIs openweb programmers techcommunity
APIs openweb programmers techcommunity
Interview success isn’t just about solving—explaining why sliding window works counts just as much. Next time, try narrating your logic as you code. Encourages clarity! Read more: https://iam.slys.dev/p/crack-google-interview-can-you-spot
🧠 Crack Google Interview: Can you spot the longest unique substring?
Google LeetCode
iam.slys.dev
January 22, 2026 at 3:16 PM
Interview success isn’t just about solving—explaining why sliding window works counts just as much. Next time, try narrating your logic as you code. Encourages clarity! Read more: https://iam.slys.dev/p/crack-google-interview-can-you-spot