Kartik

@akshen121 congo bhai

Got my first Schengen business visa and on my way for my first ever international trip🥹 Going to Germany - Europe for a business trip. A few months ago this wasn’t even in my thought. Scared, happy, proud, overwhelmed. Thankful for everything that has come my way 👍🏻

Me realising you can get frontier performance on the same model

everyone is talking about the `Agent Harness` right now. this is a great breakdown of why: → Frontier models are converging and getting way cheaper. → the real moat is the scaffolding around the LLM (context, tools, sandboxing etc.) 👇

It is good to be lucky, as long as you know it is luck and not skill.

conversely, i think most teams probably shouldn’t be building their own harness. it is unlikely you will have novel ideas around sub-agent orchestration, compaction, progressive disclosure etc that are worth owning the entire harness spend your time investing in the pieces around the harness: - execution infrastructure - custom tools, MCPs and skills - self improvement on trajectories

Will the harness be the new AI OS, in the same way Linux and Windows became the standard layer of abstraction for apps? My guess is yes! At first, we will see many harnesses, but we will slowly consolidate toward a few: one per major lab and a dominant open-source version.

@mikeaverto Handoffs? Meaning guardrails right?

@code_kartik what breaks first in an agent harness: evals or handoffs?

@RealityPolice1 @hwchase17 You will need evals for agent loop at the very least to see if you are getting expected results

@code_kartik @hwchase17 Are evals part of the harness? A harness is a runtime thing?!

@fardeentwt sorry ser

@code_kartik so big?

there are so many things i want to discuss so many takes which i want to post but this account is so big that if i post something useless good people who have followed me will be dissappointed in me and i don't want that to happen. so expect less shitposting from this account

i think i should write more articles ig

@danyay ooh i tried codex, not for me using claude code since then.

@code_kartik How much of Atlas code is now built into Browser Use plugin in codex? Maybe it wasn’t all for nothing.

Does anyone remember ChatGPT Atlas? Seems like a failed product to me. Never installed it and never even tried it.

@m31uk3 yup you are in the right dir

@code_kartik I’d build a different harness for a mobile surface than I would a desktop surface. Maybe I’m missing something and the harness is surface agnostic. 3 distinct harness archetypes though… maybe I’m right? -the claw/Hermes -the codex/Claude -the API wrapper/CoWork -the {mobile}?

🤌

@Muskanjain0401 @smallest_AI Congratulations 🎉

Life update: joining @smallest_AI to lead content series we're rolling out. 📹 smallest is building the voice layer for the next decade of AI. small, specialized, blazing fast models that quietly out-ship the giants. the thesis is simple: voice AI is a stack of small specialized models, each really good at one job, handing the conversation between them in real time, they call it artificial special intelligence. it's already the side that's winning. the stack: → Lightning → Pulse → Electron → Hydra → Atoms the receipts so far: 10M+ calls automated, 3x conversion lift, $12M in operational leakage saved, 37% lower cost per call. RingCentral, Five9, Paytm, and IDFC FIRST already on the platform. I'm here to help tell that story including deep dives, real use cases, founder conversations, benchmark breakdowns, and a lot of voice agents in action. a lot dropping soon.

So it's been around 4 weeks since I'm working on something crazy and still haven't even marketed it which even attracted a PR from @mvanhorn yesterday I'm building uWestJS v2.0 a complete platform adapter that brings uWebSockets (one of the fastest HTTP/WS implementations) to NestJS. We're talking 3-4x faster than Express (the benchmarks say itself) while keeping all the NestJS patterns you love. Basically a deep dive into systems level web programming. What I've shipped so far: HTTP Platform: - Hybrid streaming body parser (buffers small bodies, streams large ones) - Streaming compression (gzip/brotli/deflate) with content negotiation - Static file serving with ETag, range requests, and conditional GET - CORS handler with RFC-compliant preflight validation - Multipart/form-data with streaming file uploads - Full middleware pipeline (Guards, Pipes, Filters, Interceptors) WebSocket Platform: - Room-based broadcasting with efficient message distribution - Per-message deflate compression - Automatic backpressure handling (pauses when clients are slow) - Full NestJS decorator support (@SubscribeMessage, @MessageBody, etc.) - Lifecycle hooks (onConnection, onDisconnect) - Exception filters for WebSocket errors Architecture deep dive: Memory optimizations: - Stack-allocated ArrayBuffer handling - Header caching (parse once, cache forever per request) - Parsed body caching (parse once, return cached result for subsequent calls) - Cookie parsing cache (parse on first access, cache the Map) Streaming strategies: - Hybrid body parser: 3 modes (awaiting → buffering → streaming) - Automatic mode switching based on consumption pattern - Zero-copy streaming where possible (pipe directly to destination) - Backpressure propagation (pause uWS when Node.js stream is full) - Streaming compression (compress chunks on-the-fly, no buffering) Performance patterns: - Lazy initialization (only parse what you actually use) - Referential equality for empty bodies (same object reference = zero allocations) - Efficient room management (Set-based lookups, O(1) operations) - Message queue for backpressure (buffer messages when client is slow) - Abort multiplexing (multiple handlers on single uWS abort callback) Edge case handling: - Dual rejection mechanism - Graceful degradation (skip unknown compression encodings) - Defensive programming (handle non-RFC-compliant headers) - Race condition prevention Building this taught me more about core HTTP, WebSockets, and Node.js internals than any tutorial ever could. Every feature required reading specs, understanding edge cases, and making architectural decisions about memory vs speed tradeoffs. The specs/docs rabbit hole I went down: - uWebSockets.js documentation - Understanding their zero-copy architecture, backpressure handling, and why they're so fast. Learning about their streaming model and how it differs from Node.js streams. - RFC 7230 (HTTP/1.1 Message Syntax) - Implementing proper transfer-encoding handling. Did you know chunked MUST be the final encoding? Our implementation handles non-compliant headers gracefully. - RFC 7231 (HTTP Semantics) - Building CORS preflight validation that actually follows the spec. - RFC 6265 (HTTP State Management) - Cookie parsing, serialization, and security. Implementing signed cookies, httpOnly flags, and SameSite attributes correctly. - RFC 1952 (GZIP) & RFC 7932 (Brotli) - For streaming compression. We compress responses on-the-fly without buffering - crucial for large payloads. - Express.js source code - Spent days understanding their middleware patterns, body parsing, and error handling to maintain API compatibility. Your Express code should just work. - NestJS docs and adapter contracts - Deep dive into AbstractHttpAdapter, AbstractWsAdapter, ModuleRef, and the entire DI system. Understanding how Guards, Pipes, and Filters get resolved and executed. - Socket.IO internals - Replicating their room management and broadcasting API. Understanding how they handle namespaces, acknowledgments, and connection lifecycle. - Busboy documentation - Implementing RFC-compliant multipart/form-data parsing with streaming file uploads and proper backpressure handling. Launching v2.0 on NPM and publicly soon. GitHub: github.com/FOSSFORGE/uWes…

I was able to crack multiple international offers as a fresher with a cgpa below 7 and leetcode count of 38. Here is How you can do it to :

@speedyblinder Someone gets me 😂

@code_kartik Trust me the guys on X are trying to decrease competition

Dm: I am a mern stack developer and have done 600+ leetcode questions can I get a referral?