Neel Samadder

Memory Inception: The Identity Layer After a year building AI agents, I kept hitting the same wall. They look brilliant… until they restart. Every new run starts from zero. All reasoning, decisions, and progress — gone. The biggest limitation isn’t intelligence. It’s identity. Thread 🧵 #AIAgents #PersistentMemory #AgenticAI #BuildInPublic

This is what I mean when I say that I sleep like a baby.

The productivity leverage given my these agents is unfathomable but it comes at a cost where we need to evolve to consume information at such unprecedented rate and not just process but keep up and direct it The ones who develop this faster will be the ones taking over this wave #agents #ai

@shortsqueeznews Let’s talk

BREAKING: VC firm Hummingbird is raising $800 million to back "unemployable founders" who are too intense and may be overlooked by other investors. Hummingbird was an early backer of Kraken and Lovable, and all of their funds have returned 10x (solid VC performance is 3x).

@StephNass Map

Google Maps for Venture Capital. Founders and VCs, this is for you: We built OpenMap. A visual directory of 16,000+ VCs, angels, and family offices worldwide. Think of it as Google Maps for Venture Capital. 👉 Comment "map" for free access Here's what's inside: → Visual map of investors actively looking for deals → Search "investors in Paris," "deeptech VCs," "family offices" → Find intros through LinkedIn + Gmail connections → Submit your deck directly to investors If you're a VC, you should be on this map. If you're a founder, you need to know this map. 👉 Comment "map" I'll DM you the link. Make sure to follow me to receive my DM.

@Techstars Quant trading agency for boutique and retails funds

Pitch us your startup in 1 sentence. 👀

Amazing that we are getting hold of voice agents Meesho yet again proved it’s a technology company first … The future of consumerism is you speaking your mind feely and technology will pave the way by breaking the barrier it once created! #meesho #ai #agents

@akoratana @matei_zaharia @pbailis @rauchg @zoink @drewhouston 100x

Introducing: PlayerZero The world's first Engineering World Model that puts debugging, fixing, and testing your code on autopilot. We've raised $20M from Foundation Capital, @matei_zaharia (Databricks), @pbailis (Workday), @rauchg (Vercel), @zoink (Figma), @drewhouston (Dropbox), and more PlayerZero frees up 30% of your engineering bandwidth by: 1.⁠ ⁠Finding the root cause for bugs & incidents in minutes that engineering teams take days to identify. 2.⁠ ⁠Predicting in minutes, edge case issues that a 300-person QA team would take weeks to find. ------ Here's why this matters: No one in your org has a complete picture of how your production software actually behaves. Support sees tickets. SRE sees infra. Dev sees code. Each team builds their own fragmented view - and none of these systems talk to each other. When something breaks, everyone scrambles to stitch the picture together by hand. PlayerZero connects all of it into a single context graph - → The Slack thread where your lead said "we went with X because Y fell apart in prod last time" → The PR review where an engineer explained the tradeoff → The lifetime history of your CI/CD pipeline, observability stack, incidents, and support tickets So you can trace any problem to its root cause across every silo. And it compounds. Every incident diagnosed teaches the model something new. The longer it runs, the deeper it understands - which code paths are high-risk, which configurations are fragile, which changes tend to break which customer flows. So when you sit down to debug a live issue, you have your entire org's collective reasoning and production memory behind you - instantly. ------ Zuora, Georgia-Pacific, and Nylas have reduced resolution time by 90% and caught 95% of breaking changes and freeing an average of $30M in engineering bandwidth. ------ Our guarantee: If we can't increase your engineering bandwidth by at least 20% within one week, we'll donate $10,000 to an open-source project of your choice. Book a demo - bit.ly/3NlLMeN

Top quality 🤌

Who did this… epic stuff 😂😂 Explorer will be Alexandr #codex #openai @thsottiaux

token withdrawal syndrome

@grok could you please validate this an confirm the claims made ! Also could you help me understand the spare context implementation is a method ? so will it be easier for agents and foundational model labs to adapt and include this What are the barriers that still remains ? As codex is open sourced can this be implemented in codex right away? Please give a details structure answer with emphasis on all the questions raised and add pointer incase there are some major insights I missed. #Experiment #grok

@grok could you please validate this an confirm the claims made ! Also could you help me understand the spare context implementation is a method ? so will it be easier for agents and foundational model labs to adapt and include this What are the barriers that still remains ? As codex is open sourced can this be implemented in codex right away? Please give a details structure answer with emphasis on all the questions raised and add pointer incase there are some major insights I missed.

Whoa... 100M context window, with less than 9% accuracy drop Huge if true github.com/EverMind-AI/MSA

@Andrey__HQ Are you open sourcing it or could I see access to read and contribute ?

1) Agent Sentinel - threat monitoring > watches agent behavior in real-time for anomalies, prompt injection attempts, and unauthorized actions. > full audit logging so I can trace exactly what went wrong (and when)

In the last month I've built 14 internal repos for building the internet for agents. Here's exactly what each repo is and how I use it (🧵):

@superscribeio Absolutely highest accuracy with least token consumed

one issue per session is one of the easiest ways to keep coding tools useful and cheap x.com/SamadderSignal…

I did an analysis on my interactions with Codex using Codex! Key findings: - Cached inputs are the major token consumer - 85% of the cost was from cached and discounted tokens! - Using a product life cycle management tool is a must - One issue at a time gives better context, better evaluation and better results - Over-engineering and long continued sessions are major cost drivers! #AI #optimisation #tokens

The Irony of the world is the most profound and simple question are the one that still remains unanswered … The age of AI will be turning this around! #AI

The Future of Performance would be how many token of intelligence were you able to use and what throughput it gave !! #AI_era

@nikitabier So can we have feature like if one creates different folders in bookmark section! Grok can spin up and send the mess from all bookmarks section to respective folders created This would be really helpful

@slow_developer I use 5.4 high for orchestration, validation and deligation And 5.4 mini xhigh for execution and subagentic work and testing Seeing quite good results

gpt-5.4-mini with high reasoning is not good for coding, especially anything that needs spatial reasoning, like 3d geometry, and it wasn't close to gemini 3 flash, and costs more too i think its big advantage is speed. for agentic tasks, it's much faster than gemini 3 flash