Siddharth Pradhan

@YuvrajS9886 @claudeai @coderabbitai vs @claudeai

@claudeai @lamsiddharth

Introducing Code Review, a new feature for Claude Code. When a PR opens, Claude dispatches a team of agents to hunt for bugs.

Building real-time voice apps is humbling the networking layer is just as tricky as the AI parts. There's still a lot I want to understand about how WebRTC actually works under the hood. Will share what I find 👀 #WebRTC #VoiceAI #BuildInPublic #Networking #LearningInPublic

There was a second issue hiding behind the first one too. Browsers block microphone and camera access on any page that isn't HTTPS (or localhost). I was on plain HTTP from the VM's IP so the mic was completely locked out. No prompt, no warning. #WebRTC #VoiceAI #Networking

Spent way too long debugging this today I was building a voice agent using #Pipecat and SmallWebRTC on a remote VM. Tried to access it the usual way SSH port forwarding from my IDE. The connection never established. No error. Just silence. #WebRTC #VoiceAI #Pipecat #Networking

Building this requires balancing concurrency, audio buffers, and network hops all at once. It’s a reminder that great voice UX is about managing the space between sound and silence. Pipecat handles this beautifully. #VoiceAI #SoftwareEngineering #Pipecat

The magic is in the pipeline: • Streaming audio frames • Tuning VAD for precision timing • Handling interruptions • Reducing server & network latency It’s about managing the flow of sound and silence. #VoiceAI #SoftwareEngineering

Real-time voice AI is a systems engineering challenge. Diving into #Pipecat has highlighted how much work goes into the orchestration of a live conversation. You’re essentially building a real-time #distributedSystem where every millisecond counts. #VoiceAI #SoftwareEngineering

@YuvrajS9886 @alexocheema @exolabs 💯 agreed

@alexocheema @exolabs I am telling you everyone, Privacy is MOAT in the upcoming years and on device LLMs and stuff like exolabs will BLOOM!

China is way ahead on AI adoption. A school in Beijing has repurposed old macs to run personalised AI agents 100% locally using @exolabs The macs were previously used in their film studies lab, for video editing. They have ingested their entire corpus of school data: curriculums, reports, instructional materials and learning objectives - so it’s grounded to all their data in realtime. In order to get accurate answers, they need frontier models, which are BIG - memory is the constraint (not FLOPS). Apple devices with unified memory have a lot of high speed memory so stacking enough of them makes it possible to run massive models. A big concern of schools and parents is data privacy - when students or teachers use models in the cloud they are sending all their data in plaintext to the model provider. Even if schools have policies around this there’s always the risk someone accidentally copy-pastes sensitive data into the model - data leakage is inevitable.

Tried using LangGraph as a state manager for a voice AI assistant today. Instead of messy prompt chains, the workflow is modeled as a stateful graph where nodes update shared context and control the conversation flow. Feels more like building an AI workflow engine

@ethanlim siddharth.pradhn05@gmail.com

We're hiring Software Engineers at Action State Labs. $60+/hr. Fully remote. Competitive equity. Like this post and drop your email below. We'll reach out.

@sandislonjsak @Abhinavstwt Clerk was using GCP and went down for an hour, still no one noticed

AWS outage - 20 Oct 2025 Azure outage - 29 Oct 2025 Cloudflare outage - 18 Nov 2025 who’s next?

@Ai_here202 Resume

This Resume has an ATS score of more than 92🤯 This Resume helped many in getting an interview calls from companies like Google, Microsoft, Amazon, and many more. I have personally used this single-column resume in my job hunting and got amazing results I am sharing the exact similar editable ATS Friendly Resumes templates! To get it: 1. Follow me (So that I can DM) 2. Like & Repost 3. Reply "Resume" Follow me so I will dm immediately 💯

@LetMeCodeee Cpp

C++ can be confusing sometimes, but it is the most used language in HFTs and big Trading firms. Having an experience of over 3 HFTs and a work experience of over 5+ years, I have compiled all the notes and CheatSheets and Tips and tricks on A Drive. Comment “CPP” and check DM.

@bitfunded 29

Pick your lucky number from 1–50. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 1 numbers hide a surprise! 1 Winners picked will be picked randomly in 24 hours. Comment your pick 👇🏆

Follow-up to my past prisma adventures problem solved (sort of) Deleted all migration history made a fresh base migration manually marked everything as applied Prisma’s happy again. But is there no safe way to do this? #Prisma #PostgreSQL #Database #ORM

Thinking of emptying migrations table(used by Prisma to track migrations) DB is already downhill Might as well break it properly and see what happens

Crazy part Prisma stores hashes of every migration right in the DB Feels like my DB is keeping receipts for all my sins

Sometimes we need to make quick manual changes to the DB Feels harmless until Prisma goes: Wait something doesn’t add up, Want me to delete all the data? 😅 #Prisma #Database

@ptsi @kylejeong Cheap?? Gemini computer-use is expensive

@kylejeong Smarter, cheaper, or faster… pick all 3? :)

The new Computer-Use model from @GoogleDeepMind has been crushing it in trusted benchmarks, We benchmarked the model against the new models from @AnthropicAI as well as @OpenAI on Online Mind2Web & Web Voyager. Gemini crushes them all in accuracy, speed, and cost (by a lot).