Aadil Manazir

14 posts

Aadil Manazir

Aadil Manazir

@0xaadilm

Katılım Eylül 2022
926 Takip Edilen91 Takipçiler
Aadil Manazir retweetledi
Winston Weinberg
Winston Weinberg@winstonweinberg·
We had an incredible April at Harvey. - Net new ARR is up 6x YoY - We’re about to break 50% DAU/MAU - Our average user now spends 12 hours a month using Harvey Job's not finished.
Winston Weinberg tweet media
English
25
33
429
167.1K
connor ling
connor ling@conconconling·
the founding engineer math today is entirely broken -- compared to options like joining one of the top frontier labs that seem to be on track for world domination and founding one's own company leveraging the abundant early-stage funding market, getting 1% working for someone else is understandably unappealing founders must create stronger / more novel incentive structures to have even a shot to win in the talent war: raise less dilutive early funding rounds and reserve more equity for hires, offer commissions / profit shares for customer contracts, offer significantly above-market equity to attract mission-aligned contributors...
English
22
9
400
116.5K
Aadil Manazir
Aadil Manazir@0xaadilm·
Excited to share what I've been working on! I designed and led Harvey's new document drafting agent from zero to GA. It now powers docx editing across all Harvey's products. Grateful to work with awesome teammates across our Embedded Experience and Assistant teams!
Harvey@harvey

Pushing the Harvey for Word Add-In further exposed the limits of .docx. So Harvey’s engineers designed an agent that builds documents in memory, offloading state to the backend so the model can focus on legal text. 40% higher edit acceptance. 70% more usage.

English
1
2
7
395
Catherine
Catherine@0catd·
The way we think has already been shaped by strangers' design decisions. I wrote about how chat interfaces are the new 'water we swim in,' hiding how meaning gets lost every time we compress a thought into a text box. We shape our tools, and then our tools shape us. Read below👇
English
2
0
4
190
Catherine
Catherine@0catd·
New design experiment I made for a layered ai chat interface! The current "wall-of-text" output makes information hard to process, so I built an interface that layers responses into collapsible sections. Read the full blog post on my site.
English
2
0
4
180
samarth @ ICLR
samarth @ ICLR@samarth__go·
Excited to share what I've been building recently! We built a multi-agent pipeline to source, validate, and productionize legal infrastructure - scaling @harvey's knowledge sources from 6 to 60+ jurisdictions. Looking forward to pushing both coverage and quality even higher!
Harvey@harvey

In legal work, outcomes depend on the quality and coverage of your knowledge sources. In this blog, @samarth__go and Christopher Bello explain how we built the Data Factory to discover authoritative legal sources, validate them for compliance, and test real legal reasoning at scale. The result: Harvey scaled from six jurisdictions to 60+, and from 20 legal data sources to 400+. Full breakdown: harvey.ai/blog/using-age…

English
4
1
16
3.2K
Aadil Manazir
Aadil Manazir@0xaadilm·
@kashyechuri I always update when the popup shows up. I'm on 2.3.34 Am I supposed to see a chat tab here?
Aadil Manazir tweet media
English
0
0
1
62
kash
kash@kashyechuri·
@0xaadilm what version are you on? we added a bunch of stability improvements in 2.3. let me know if you’re still having issues!
English
1
0
2
143
kash
kash@kashyechuri·
Guide to getting the most out of coding agents (from the Cursor team): - plan before coding. when agent runs don't match what you wanted, revert changes and go back to the plan - let the agent find context, you don’t need to manually tag every file - start new conversations when switching tasks or if the agent seems confused - @ past chats instead of copy pasting whole conversations - rules = static context that applies to every conversation. start simple. add rules only when you notice the agent making the same mistake repeatedly. - skills = workflows agents can invoke when needed. you can set up hooks that keep agents grinding until all tests pass (see full guide below) - agents can process images directly from prompts (screenshots, design files, image paths). you can scribble on a piece of paper then share with Cursor - review review review - review during generation with the diff view - after the agent finishes, click review → find issues to run a dedicated review pass - use Bugbot for automated review on your PRs - for significant changes, ask the agent to generate architecture diagrams - run multiple agents in parallel, Cursor automatically creates and manages git worktrees for parallel agents. you can also run the same prompt across multiple models simultaneously - delegate tasks you’d add to a todo list to cloud agents (trigger agents from Slack with “@ Cursor”) - use debug mode when standard agent interactions struggle with a bug. provide detailed context about how to reproduce the issue. the more specific you are, the more useful instrumentation the agent adds. - devs who get the most from agents share a few traits: write specific prompts, iterate on their setup, review carefully, give agent verifiable goals, treat agents as capable collaborators
English
20
13
238
18.7K
Aadil Manazir
Aadil Manazir@0xaadilm·
+1, this is my current workflow I don't know how people use the CC extension when it doesn't even have syntax highlighting. Half the time I'm using Cursor to understand what's going on, and I can't even read the CC extension results. Cursor UX is way better compared to CC terminal and extension
English
0
0
2
155
sam
sam@samgoodwin89·
Maybe I'm a dumbass noob, but I can't function without all of this on one page. Cursor is still my main driver. Git changes on left - click to see diffs. Code is next, with diff overlays showing what the agent did. Chat. Agents list. What am I missing about these CLI tools?
sam tweet media
English
147
8
366
75K
Catherine
Catherine@0catd·
Back in February, I prototyped this feature and other novel ones like visual representations of trains of thought, a multipane chat interface, and an infinite canvas chat. I wrote about my process and the history of these interfaces here: catherinedavodi.com/branch-chat-wr…!
OpenAI@OpenAI

By popular request: you can now branch conversations in ChatGPT, letting you more easily explore different directions without losing your original thread. Available now to logged-in users on web.

English
2
0
7
1.2K
Aadil Manazir
Aadil Manazir@0xaadilm·
Just eat more high calorie food You’re fine on total protein amount Full fat Greek yogurt, eat 1/2 a tub - a tub a day Make a protein smoothie, milk, peanut butter, berries, protein powder, etc Up your carbs: 3-4 slices of whole wheat bread, a few cups of Cheerios Try eating for most of the day while you’re working. Source: I also have a hard time gaining weight. Started at 6’3”, 130 lb, bulked to 6’3”, 195 lb, about 15-16% body fat
English
0
0
3
223
gabriel
gabriel@gabriel1·
I NEED ADVICE, i can't seem to gain weight. just measured myself again and i lost weight, even after heavy lifting and eating 1 pound chicken / salmon per day + nuts in morning, im 6'1" and just measured 152lbs
English
1.2K
14
1.5K
678.4K
CTRL Sheet
CTRL Sheet@ctrlsheet_ai·
🚀🎉 CTRL Sheet alpha is launching today! Meet CTRL Sheet, your AI agent for spreadsheets built directly into Google Sheets and Microsoft Excel! Start with our alpha version on our website, and get ready for upcoming extensions for Excel and Google Sheets! ✨ CTRL Sheet can: ✅ Build models and tables ✅ Edit cells effortlessly ✅ Reference external CSVs/Excel files ✅ Create/edit charts ✅ Fetch real-time web data ✅ Much more! Watch our demo video below!
English
3
2
25
4.4K
MegaMafia
MegaMafia@megamafia·
Are you autistic or are you crypto autistic
English
1
0
3
748