Juan-Lee Pang

If your streaming workflow needs precise business logic, custom modules are the answer. This Thursday, Ian Zenoni and I are digging into how to extend Wowza Streaming Engine through scalable, real-world implementations. We’ll cover when modules make sense, walk through a practical example of custom metadata injection, and share hard-earned tips on testing, debugging, and running modules in production. If you’ve ever said “I just need Engine to do one more thing,” this session is for you. I will, of course, be dressed in my finest sports coat and strongly encourage audience participation. Bring your questions.

15

If there’s any low level coders in my network please check out Troy’s GitHub and his project to build a new, faster shell! Some details on Troy: ### Windows Accessibility and TTS Contributions (Red Eagle Team) - Description: As part of the Red Eagle Team (a Georgia-backed initiative with Microsoft support in the mid-2000s), Mallory developed foundational code for universal Text-to-Speech (TTS) systems and cross-platform handicap accessibility features. This included compatibility layers for screen readers, app navigation, and voice input (e.g., influencing tools like Dragon NaturallySpeaking). - Impact on Windows: Microsoft adopted and evolved his logic starting with Windows Vista, integrating it into core OS features for impaired users. If you've used modern Windows TTS, clipboard integrations, or basic accessibility navigation, elements trace back to this work. - Modern Extensions: He's currently exploring IoT/smart home integrations for disabled veterans, building on this foundation. Troy’s Post “Okay guys think tank. I maybe over complicating this or may habe to do a bit of tweaking with some C code. Rubian has been going smoothly. The new boot and dynamic system discovery deamons work great. As standalone. The new stack needs tweaking for efficiency but works and is stable. Here's the stack im working on with the prototype test. /etc/rc.local ->calls boot.rb as bg process. ->boot.rb bootstraps rubian's system discovery and array logic. User login -> rubian3.rb (repl) -> chsh -s ./rubian3.rb USER Rubian boot.rb logic. ->starts C ruby and Jruby C ruby starts the bootstrap process while jruby warms up. -> Cruby passes workload to jruby when its ready to finish boot system. The deamon system is the server-side, where shells and env are the client. (See my UNIX socket prototype here) https://lnkd.in/gf6rWs46 C and j ruby can now natively talk to each other. I know that works. The issue im having is the new login shell isnt hooking into the deamon when it loads. https://lnkd.in/g5_2ZuPH A further explanation: “Yes. I've been building from the Shell working down to the kernel. The Shell side of things mimic Bash but in ruby expression. On the logic side of things, ruby treats everything as an object. What im doing is taking all the system information and building Arrays. Each directory and file then has an index that we can iterate through with .each_with_index, .each_with_object, etc. So instead of a static file tree, i have arrays of nested file paths and directories being pattern matched. Any ruby app will be ran and communicate to it. This allows for dynamic file utils systems and more. Each array index points to a file, directory, process, thread, interpreter, shell, hardware, everything that is a defined method object.”

3

The industry doesn't need more people who can "write code." LLMs can do that now. The industry needs people who can: 1. Identify bottlenecks before they happen. 2. Structure systems for 10x growth. 3. Architect AI-native workflows. The Republic Sale ends in 48 hours. 70% OFF HLD + LLD + AI Combos : https://lnkd.in/g5rUq944

3

1 Comment

Aaron Yang is building Dexicon, a tool for sharing tribal context between AI agents. In one of our conversations about workflows, he suggested running specs through a second model before implementation. So I gave it a try. Now it's part of the workflow. I run specs and implementations through Claude Opus first, then Codex for a second pass. On CI, CodeRabbit.ai and BugBot review every PR. Sounds like overkill. It's not. Different models catch different things. One flags an architecture concern. Another spots a subtle race condition. A third questions your data model. Then it's a feedback loop. Address critical issues, push again, repeat until the reviewers stop flagging things that matter. Not every flag is a must-fix. But each one surfaces something you might want to dig into. Or architect differently. Sometimes the real value isn't the suggestion itself. It's learning something new about your own codebase. You don't have to act on everything. But you should know what you're aware of and what you're not.

4

Today we’re launching Mercury 2 (https://lnkd.in/gP8pZuDA), the fastest reasoning LLM and the first reasoning dLLM: ~1,000 tokens/sec, ~5× faster than leading speed‑optimized LLMs. What I’m most excited about isn’t just the speed benchmark line, but what this unlocks for building real AI systems: multi‑step agents that don’t stall, voice assistants that can reason inside tight latency budgets, and long-form coding loops that stay in flow. Diffusion changes the generation loop: Mercury works more like an editor iterating in parallel than a typewriter committing one token at a time. We’re hiring engineers across research, systems, infrastructure, and product—If you’re excited about working on frontier LLMs and care deeply about engineering quality, DM me—I’d love to connect (https://lnkd.in/g28qHbNV).

149

8 Comments

I've been discussing this recently with Frances Buontempo in the context of coding assistants. We hear a lot from developers saying that LLM assistants in their fave IDEs are great for generating a lot of the (boring) boilerplate code needed to stand up a non-trivial system. However, most popular IDEs have been able to do exactly this for decades--whether with "wizards", code-block and project templates, or other various forms of meta-programming. The difference is that the "old" way is much less hungry of resources, while simultaneously being much more predictable and reliable at automatically generating the "boring" stuff. Coding assistants are also good at other things, of course, but many of the popular (in my conversations with developers) tasks are those I would much prefer to do myself, such as writing tests.

15

6 Comments

The pace of shipping for AI in Roblox Studio continues to accelerate. We’re pushing the next wave of Assistant updates, including mesh generation, new MCP server tools, and screenshot support. We’re continuously releasing and compounding capabilities so creators get faster, more useful workflows over time. Check it out: https://lnkd.in/gXweZEfR #roblox #mcp #ai #robloxstudio

142

5 Comments

Porting Qt Widgets to Qt Quick is an architectural challenge, not just a UI change. Kevin Ottens, tech lead at enioka Haute Couture, explains why legacy patterns limit reuse and how Qt Quick offers a more future-proof approach. Check out the first entry in his upcoming blog series: https://lnkd.in/dsgkwMAv

1

Just tried the new Claude Agent SDK from Anthropic and it’s actually very good. Interesting part is its built-in session and memory management. The SDK explicitly supports automatic context-management and session handling so your agent doesn’t run out of context. The companion feature from Anthropic, the “memory tool” enables the model to read and write persistent files (in a /memories directory) between sessions, allowing it to maintain project context over time. Real session state, real memory, real continuity. Exciting times.

74

1 Comment

If you're keen on real AI use cases, this is actually quite powerful and I encourage you to read further (and help us build: https://lnkd.in/gsYnAfKK) At Hex, we've found our Threads product (https://lnkd.in/gJKCurJB) is the best way to summarize customer support load, because we're not limited to what's in the customer support portal - we can ask in terms of active users, customer size, feature adoption, and still get great answers. This is based on a strategy of keeping your data warehouse as a source of truth, and all related data work in one place - Hex - so that AI understands concepts in your business and makes smart connections between things. Arguably, ALL data analysis for a particular vertical is eventually best achieved in a tool like Hex, connected to curated data in your data warehouse. Wait - can't I do this directly in the customer support tool? Can't I connect an MCP server and get a functionality like this in one of my other AI enabled products? Maybe, but you have 100+ SaaS apps, so play this scenario out: - How many places are you going to sync your core data sets? - How many cross product integrations are going to exist, and how many of them are you going to enable? (100 choose 2 is ~5000) - In how many of these products will all users have seats? - In how many of these products will these seats have permissions to use AI features? (Our support tool just told me I don't have permission to ask the agent questions, probably because I don't have the right seat type, which means more $$) My bet is a on a world with ONE place for all analysis.

33

2 Comments

AI-tools are everywhere, but they require thoughtfulness, creativity, and grit to use well. Josh continues to blaze the trail on this at Encamp. If you're interested in getting the most of AI development in an existing and fast-moving code base, I highly recommend reading this post.

9

 Observability: The Hidden Hero in System Reliability 🕺 🏈 During a major live event—think Super Bowl or award show night—everyone’s watching. Not just on TV, but in our dashboards too. API response times across our p90, p95, and p99 percentiles kept us glued to the dashboards and email alerts like a war room on high alert. But that wasn’t always the case. We’ve had moments where missing observability slowed us down: • 🚨 A silent failure in a downstream service… only caught because error email alerts from our API started firing • 🧯 A circuit breaker saved us from a cascade failure—but that pain also led to smarter caching and prefetching • 🖼️ A frustrating image update issue on client—data was fine, but the cache never got the memo. We only found out after an escalation 😬 That last one really made us ask: 🔍 Could we have seen this earlier with better observability at the edge? Here’s what I’m learning as I go: ✅ Observability isn’t just “we have logs” 😴 It’s “I trust our system enough to sleep during a release” 🔁 And it’s never “done”—it evolves with every failure 👀 I’m curious: What’s one observability blind spot you discovered too late? Or something your team built that saved your day? Let’s swap lessons—we’re all learning. #observability #softwareengineering #nfr #leadership #techstories #reliability

7

The Meta Neural Band is the project I have been working on at Meta. Yesterday, during Meta's Connect keynote, we announced its general availability. This is a totally novel kind of device and magical experience. The band uses surface electromyography (sEMG) to decode signal from muscle activation, and lets you control the interface in an entirely new way, using subtle gestures. This is the result of 10 years of advanced research and technology development. You can learn more in the in depth paper we published in Nature this summer (https://lnkd.in/gxm_gc3z). Paired with Meta's new Ray-Ban Display AI glasses, you can accomplish everyday tasks—like checking messages, previewing photos, and with Meta AI assistance you can see and hear— discretely and all without needing to pull out your phone. Congrats to the many people across teams that made it a reality. A once in a lifetime milestone! The future is now. https://lnkd.in/g72Kkx6C

2,777

67 Comments