Anuar

551 posts

Anuar banner
Anuar

Anuar

@_startuphacker

CTO @ https://t.co/73ZP71gNDB - AI agent for documents | my journey from Kazakhstan's steppes to building AI B2B SaaS

Toronto, Ontario شامل ہوئے Temmuz 2017
217 فالونگ6K فالوورز
پن کیا گیا ٹویٹ
Anuar
Anuar@_startuphacker·
We just launched! And we are #1 on ProductHunt! 🚀 Tyce is an AI Agent that crafts the smartest documents. With Tyce you can write in the flow, Cursor-like vibes but for knowledge workers. Tyce is deeply integrated with our Word Processor, it can find relevant information in your company’s file storage and generate polished documents in seconds. After talking to over 300 people to learn their processes and pain points - sales, operations, compliance and legal, we were surprised to learn that despite advancements in LLMs people still do document workflows manually. From finding the relevant sources, drafting the first version to reviewing and eventually signing documents - this workflow takes weeks, sometimes months! Existing tools like Microsoft Copilot and Drive Gemini have little adoption and sub-par UX, we realized that one needs to rebuild the whole UX and backend from the ground up so that you can write in the flow. We want to reimagine productivity tools with AI, starting with the most popular productivity tool - Document Processor.
English
36
10
338
43.4K
Anuar
Anuar@_startuphacker·
Maybe the new wave of startups will be consulting shops that teach how to use Claude Code 🙃
Todd Saunders@toddsaunders

I know Silicon Valley startups don't want to hear this..... But the combination of someone in the trades with deep domain expertise and Claude Code will run circles around your generic software. I talked to Cory LaChance this morning, a mechanical engineer in industrial piping construction in Houston. He normally works with chemical plants and refineries, but now he also works with the terminal He reached out in a DM a few days ago and I was so fired up by his story, I asked him if we could record the conversation and share it. He built a full application that industrial contractors are using every day. It reads piping isometric drawings and automatically extracts every weld count, every material spec, every commodity code. Work that took 10 minutes per drawing now takes 60 seconds. It can do 100 drawings in five minutes, saving days of time. His co-workers are all mind blown, and when he talks to them, it's like they are speaking different languages. His fabrication shop uses it daily, and he built the entire thing in 8 weeks. During those 8 weeks he also had to learn everything about Claude Code, the terminal, VS Code, everything. My favorite quote from him was when he said, "I literally did this with zero outside help other than the AI. My favorite tools are screenshots, step by step instructions and asking Claude to explain things like I'm five." Every trades worker with deep expertise and a willingness to sit down with Claude Code for a few weekends is now a potential software founder. I can't wait to meet more people like Cory.

English
1
0
2
617
Cursor
Cursor@cursor_ai·
Composer 2 is now available in Cursor.
Cursor tweet media
English
572
868
9.4K
4.7M
Anuar
Anuar@_startuphacker·
@rohanpaul_ai Imagine the opportunity Google has with GMeet as data capture, GDrive as system of record and the rest of the Workspace stack for permissioning and finally delivering final outcomes as documents or code
English
0
0
0
43
Rohan Paul
Rohan Paul@rohanpaul_ai·
Ali Ghodsi, the cofounder and CEO of Databricks, says Zoom has a massive chance to build an AI-first product, that could seriously disrupt the traditional enterprise SAAS. Because it sits on the largest datasets of meeting videos and transcripts. The big pain in enterprise software is data entry and coordination. Zoom already sits on the raw input: every customer call and internal meeting, plus the video, audio, and transcript. If Zoom can reliably pull out decisions, context, and action items, then write them back into the right system of record automatically, as an AI-first workflow layer, it becomes the front door for work. That would replace lots of separate SAAS tools that exist mainly to collect notes and updates. --- Video from 'Bg2 Pod' YT channel (link in comment)
English
49
24
281
82.2K
Anuar
Anuar@_startuphacker·
@llama_index Awesome, will try out on the weekend!
English
0
0
5
965
LlamaIndex 🦙
LlamaIndex 🦙@llama_index·
We've spent years building LlamaParse into the most accurate document parser for production AI. Along the way, we learned a lot about what fast, lightweight parsing actually looks like under the hood. Today, we're open-sourcing a light-weight core of that tech as LiteParse 🦙 It's a CLI + TS-native library for layout-aware text parsing from PDFs, Office docs, and images. Local, zero Python dependencies, and built specifically for agents and LLM pipelines. Think of it as our way of giving the community a solid starting point for document parsing: npm i -g @llamaindex/liteparse lit parse anything.pdf - preserves spatial layout (columns, tables, alignment) - built-in local OCR, or bring your own server - screenshots for multimodal LLMs - handles PDFs, office docs, images Blog: llamaindex.ai/blog/liteparse… Repo: github.com/run-llama/lite…
English
11
44
365
178.8K
Anuar
Anuar@_startuphacker·
@OpenAINewsroom @astral_sh Probably to integrate dev tooling into the LLM training process for more end-to-end training
English
0
0
1
201
OpenAI Newsroom
OpenAI Newsroom@OpenAINewsroom·
We've reached an agreement to acquire Astral. After we close, OpenAI plans for @astral_sh to join our Codex team, with a continued focus on building great tools and advancing the shared mission of making developers more productive. openai.com/index/openai-t…
English
480
821
7.2K
3.9M
Anuar
Anuar@_startuphacker·
@contextkingceo Hot take: relations can be inferred by the agent at query time
English
0
0
2
132
Nishkarsh
Nishkarsh@contextkingceo·
We've raised $6.5M to kill vector databases. Every system today retrieves context the same way: vector search that stores everything as flat embeddings and returns whatever "feels" closest. Similar, sure. Relevant? Almost never. Embeddings can’t tell a Q3 renewal clause from a Q1 termination notice if the language is close enough. A friend of mine asked his AI about a contract last week, and it returned a detailed, perfectly crafted answer pulled from a completely different client’s file. Once you’re dealing with 10M+ documents, these mix-ups happen all the time. VectorDB accuracy goes to shit. We built @hydra_db for exactly this. HydraDB builds an ontology-first context graph over your data, maps relationships between entities, understands the 'why' behind documents, and tracks how information evolves over time. So when you ask about 'Apple,' it knows you mean the company you're serving as a customer. Not the fruit. Even when a vector DB's similarity score says 0.94. More below ⬇️
English
613
661
5.9K
3.8M
Anuar
Anuar@_startuphacker·
@JonahLupton It’s an anonymous platform, someone can easily start a swarm of agents creating infinite content
English
0
0
0
80
Anuar
Anuar@_startuphacker·
@MatthewBerman They should just take open source models and integrate them in FB, would save a few hundred billion 😭
English
0
0
0
228
Matthew Berman
Matthew Berman@MatthewBerman·
How happy is Meta to not have a frontier model right now
English
100
19
997
97.7K
Anuar
Anuar@_startuphacker·
@internetvin Connect my quickbooks and we get best investor matches lol Boardy 2.0
English
0
0
1
65
internetVin
internetVin@internetvin·
Alright so we have been working on this. We want to create some kind of centralized revenue tracker for the entire Canadian technology ecosystem. This would be a number we are all connected to in some way, we can all work on it together, and see how it goes up and down. Revenue sits at the middle of a bunch of different companies and developments. It's a number that is relevant to both funded and bootstrapped companies. And it's a number that still matters even in a world where companies and individuals can do a lot more with a lot less. It's going to be very hard to bring all of this data together, so we are going to need some kind of compelling or unique incentives to get people to share this information. But it can all be anonymous. The goal is to just create a single number to rally around that is regularly maintained and updated. Let me know what you think. Thank you for everything so far.
internetVin@internetvin

Is there a well maintained central database of all technology companies in Canada? One with high quality information that’s updated on a decent cycle?

English
8
1
36
12.7K
Anuar
Anuar@_startuphacker·
@ohmypy Best use of my personal human thinking tokens is in prompt engineering and verifying, why would i write meta documentation if it’s already in the code
English
0
0
0
77
Anton Zhiyanov
Anton Zhiyanov@ohmypy·
Anyone with an LLM can generate megabytes of documentation in minutes. Nobody wants to read it, of course. Clear and concise docs with a high signal-to-noise ratio are more valuable than ever. But almost no one can actually write them. The art of writing short docs is lost.
English
35
23
378
12.3K
Anuar
Anuar@_startuphacker·
We will start “levelling” engineers where in the level of the stack you operate in: do you design and prompt engineer? Or are you low-level optimizer and verifier? The best can do both!
Michael Truell@mntruell

x.com/i/article/2026…

English
0
0
1
302
Anuar
Anuar@_startuphacker·
We find that in some specific coding domains LLMs still don’t strive yet, in our case our editor is written in Prosemirror package in javascript and we still write code by hand here. Interestingly the structure of this package is very “segregated” in a way that modules are highly decoupled and”far away” yet inter-related, this makes next token prediction to fail.
English
1
0
3
264
Anuar
Anuar@_startuphacker·
@surim0n Sign me up
English
0
0
0
321
Saurabh Suri
Saurabh Suri@surim0n·
Announcing: Claude for Everyone: Toronto We're hosting an official Claude Code for Everyone Community Meetup. Not just for developers this time. Anthropic has been shipping fast. Claude has changed how we work. Claude Cowork, Skills, and Connectors are bringing that same power to PMs, founders, ops people - anyone who works on a computer. Most people haven't seen any of this in action yet. That's what this event is for. Real demos. Hands on with the full Claude ecosystem. Space is very limited. Are you a Cowork power user and want to demo? DM me Brought to you by @buildfutureto cc: @robjama
Saurabh Suri tweet media
English
65
27
573
116.7K
Anuar
Anuar@_startuphacker·
@nicbstme Try our Agent for Word, the most flexible agent, and we will release agent for spreadsheets in the coming months too
English
0
0
0
124
Leo 🏴‍☠️
Leo 🏴‍☠️@leostera·
“Across multiple coding agents and LLMs, we find that context files tend to reduce task success rates compared to providing no repository context, while also increasing inference cost by over 20%.” arxiv.org/abs/2602.11988 👀
English
10
34
194
29K
Anuar ری ٹویٹ کیا
Citrini
Citrini@Citrini7·
JUNE 2028. The S&P is down 38% from its highs. Unemployment just printed 10.2%. Private credit is unraveling. Prime mortgages are cracking. AI didn’t disappoint. It exceeded every expectation. What happened?​​​​​​​​​​​​​​​​ citriniresearch.com/p/2028gic
English
1.9K
4.3K
27.9K
28.5M
Anuar
Anuar@_startuphacker·
@mdancho84 A mistake of every software engineer to rely on simple processing libraries like this, works for prototypes but too simple for complex structures
English
0
0
1
193
Matt Dancho (Business Science)
🚨 BREAKING: Microsoft launches a free Python library that converts ANY document to Markdown Introducing Markitdown. Let me explain. 🧵
Matt Dancho (Business Science) tweet media
English
40
141
1.3K
142.9K