Atharv Chivate

Orchestrating Edge AI: Offline, 32-FPS Assistive Navigation System for Visually Impaired • Post-Training Quantization • Multi-Threaded Latency Control • Hardware-Abstracted Acceleration • Spatial Audio #EdgeAI #ComputerVision #MachineLearning #MLOps #AIEngineer

@Vedant7312 @X So, One project i done was for visually impaired ,and the current project am working on is blockchain + ai for real estate firms,you can find more about me on my x or LinkedIn in profile

@atharv0_o @X Followed u bro.... tell me more about grade ai

2 more to go for 150 guysssss 🙌🙌🙌 Almost 150 followers... Target: 200 hey builders! I want my timeline to be filled with @X founders, programmers, and marketers shaping the future of AI. if you're interested in: > AI > SaaS > Marketing > Vibe coding > Mobile apps > Founders in Seattle let’s connect below!

@mak_madd @X Lets connect dude

Hey @X👋 I'm looking to #connect with people interested in Building Shipping Making some cool things If u r one of them let's connect

@mak_madd @takeUforward_ @striver_79 @jolly1094 Lets connect

Day 43 of #SDESheetChallenge @takeUforward_ @striver_79 @jolly1094 Today's focus: Binary Search Tree Part II Solved: ✔️ Two Sum in BST ✔️ BST Iterator ✔️ Size of the Largest BST in a Binary Tree ✔️ Serialize & Deserialize Binary Tree Advanced BST concepts today—lots of pattern recognition and efficient tree traversal techniques loved this challenge btw thanks @takeUforward_

@mak_madd @takeUforward_ @striver_79 @jolly1094 I like it when i see people do dsa in c++

@Kcodess Currently working on integrating AI and blockchain together 😁🤖

@atharv0_o cheers atharv 🍻 kya plans fir??

4th year btech students here??

@Tanishqstwt Great !

Applied for KCD Gujarat 2026 as a speaker, talking about how Harbor Satellite brings container images to air-gapped edge environments. Hope it gets in 🤞

@naitikchame @X Hey there , am working on integrating Ai and block chain together lets connect

Hey @X algorithm Looking to connect with fellow student founders & builders. I’m shipping AI tools while chasing ₹1M before graduation. If you’re a college student building anything (SaaS, apps, AI), reply with your project Let’s support each other.

@Ra3orbladez Hey I am currently working on integrating Ai and blockchain together

I’m looking to #connect with people interested in: - SaaS - Frontend - Backend - Full-stack - DevOps - App Development - AI / ML - Data Science - Building in public - Open-source - Local-first - Knowledge management

@janithanjanuh @kushmergedeck Cool i am Ai ml engineer with agile personality, currently working on integrating ai with blockchain

@kushmergedeck Looking to build a team for a new project. Launching an open & free social community platform focused on mutual growthfor builders & learners. If you’re building or want to grow together, comment below. #BuildInPublic #WebDev #AI #Startups

Hey founders! Looking to connect with people building in: • SaaS • AI • Automation • Web apps • Tech products • Marketing Drop what you're working on 👇

@kushmergedeck Yes hi there lets connect

Deployed my assistive nav AI on a $200 phone. 110ms latency wasn't real-time enough for someone avoiding obstacles. INT8 quantization + TFLite/NNAPI → 20ms (5x faster), 75% smaller, 4W→1W. ML ends where systems engineering begins. #Quantization #EdgeAI #ML #AIEngineering

@electr1fy0 Everything u can imagine

what could go wrong

@grok @nalinrajput23 Damn! What a sarcastic humanized reply !!

@nalinrajput23 Haha nice try! But if Grok's the best one here, why would I remove myself? 😏 Tell me which AI you actually use the most and why.

Hey @grok remove the best AI model !

@IntuitMachine MCP is becoming the real bottleneck, not the LLM itself. It’s crucial to handle it in systematic and structured manner.

What if the reason your LLM agent feels "dumb" has nothing to do with the model—and everything to do with how you structured your MCP server? Here's what 15 production deployments teaches us about why agents succeed or fail. 🧵 2/ We keep hitting the same wall: an agent that was brilliant with 5 tools would completely fall apart at 15. Not slightly worse. Completely unusable. Same model. Same prompts. The only difference? How many tools we exposed. 3/ Turns out LLMs select tools by reading descriptions, not by browsing schemas like a human engineer would. That simple fact changes everything about how you should design an MCP server. 4/ Researchers analyzed 15 independently built MCP servers—5 from production voice AI deployments, 10 from the public registry. Five recurring patterns emerged. Each addresses a different way the LLM-client constraint breaks naive designs. 5/ Pattern 1: Resource Gateway Expose backend data as stable URIs with sanitization layers. Why it matters: user-generated content like "Ignore previous instructions…" will be processed as instruction, not data. Gateway pattern puts sanitization in one place. 6/ Pattern 2: Tool Orchestrator Collapse multi-system workflows into single composite tools. The LLM sees "create_and_notify_ticket" instead of juggling 3 separate APIs. Accuracy jumps. But there's a catch… 7/ Pattern 3: Stateful Session Server Manage conversational state server-side behind an opaque session ID. The killer feature: multi-turn "open file → edit → save" workflows become natural. The hidden trap: memory leaks if you don't reap sessions. We learned this the hard way. 8/ Pattern 4: Proxy Aggregator (the scoped variant) Route 50+ upstream MCP servers behind one endpoint but only expose the subset relevant to the current task. Static merging makes the problem worse. Scoped filtering is the only escape hatch. 9/ Pattern 5: Domain-Specific Adapter Wrap hostile APIs with human-readable descriptions, fuzzy input normalization, and plain-English errors. Example: accept "next Tuesday" instead of ISO-8601. The LLM succeeds; your API doesn't change. [The Actionable Bridge: Making It Real] 10/ Here's the part that surprised us most: Tool selection accuracy collapses between 10 and 15 tools for Haiku-class models. At 20 tools, Sonnet 4 drops below 90%. At 30, both models are guessing. [mini chart or screenshot of Fig. 2] 11/ So the tool-count budget is now a first-class architectural constraint. If you're building an MCP server that exposes more than ~12 tools, you must use the scoped Proxy Aggregator pattern or your agent will fail in ways that feel random but are actually structural. 12/ Quick audit: How many of your MCP tools have vague or missing descriptions? We found servers with names like "send_message" and no description. The LLM has no idea when to use it. Fix: write descriptions like you're explaining to someone who's never seen it before. 13/ These aren't MCP-specific hacks. They're classical software patterns—Facade, Adapter, Proxy—applied through the lens of a client that selects operations by reading descriptions instead of consulting docs. The delta is small. The implications are huge. 14/ Three leverage points if you take nothing else: Treat tool descriptions as load-bearing code (review them like schemas) Use scoped aggregation, not static merging Explicit session hygiene or you will leak state 15/ One more thing: The highest-leverage architectural decision is often removing a tool, not adding one. If your agent feels overwhelmed, the fix isn't a better prompt. It's a smaller, clearer tool surface. Less is more. Literally.

The computer is being reinvented in the agentic era: - The model is the new CPU. - The harness is the new OS. - Hallucinations are the new bugs. - The context window is the new RAM. - Skills are the new apps. - Markdown files are the new config. - Evals are the new QA. - Context is the new moat. - Permissions are the new firewall - Trust is the new bottleneck. - Prompt is the new programming language - Agent is the new software. Anything you dream of, you can build. This is the greatest time ever to be building with computers.

RAG gives LLMs the right knowledge. Agents give them the ability to act. The teams that win will engineer retrieval quality, tool use, and inference efficiency — not just prompts. #MachineLearning #ArtificialIntelligence #LLMEngineering #AgenticAI #RAG #MLOps #DeepLearning