Connor Burke

473 posts

Connor Burke banner
Connor Burke

Connor Burke

@cmb

Online account for online things, Technology ERP Implementation Consultant, Constantly learning, Autodidact

Washington, DC Katılım Mart 2007
609 Takip Edilen4.1K Takipçiler
Sabitlenmiş Tweet
Connor Burke
Connor Burke@cmb·
New personal website inspired by Apple iOS It’s simple, interactive (for the most part) & can be navigated by clicking the screen It was a ton of fun to build during the free lovable day Give it a try and leave me some notes 👋 (I know it’s cropped) connorburke.com
English
2
0
2
317
Garcia
Garcia@GarciaCap·
“How bullish are you in Anduril?” Me:
Garcia tweet media
English
49
3
825
120.9K
Kevin Kwok
Kevin Kwok@kevinakwok·
@cmb Haha fresh from my oven
English
4
0
149
11.1K
Connor Burke
Connor Burke@cmb·
Agreed FDE 📈 Read the article and made me curious on your thoughts on joint ventures like Microsoft X Accenture to create @avanade (consulting arm only for Microsoft products) which could be similar to consulting arm focused solely for Palantir implementations. In house obviously better but curious your opinion on joint ventures with consulting companies to have a trained “Deloitte Palantir FDE Specialist” type employees
English
0
0
0
45
Connor Burke
Connor Burke@cmb·
@chadwahl @pitdesi Do you think consulting companies that offer these services will lose market share and be replaced by the FDEs at PE firms like this? Been seeing this recently the switch to internal hires deployed onsite for new acquisitions and then rotated to another site after implementation
English
1
0
1
235
Andrew Yeung
Andrew Yeung@andruyeung·
My new favorite NYC life hack: take Ubers for ~50% of the price It’s called Empower I took a car from Manhattan to Newark during peak time. It cost $61 (usually >$120) What’s the catch? I looked into it: Drivers pay a monthly subscription fee to be listed on the app. In return empower takes no cut (vs 30% from Uber) Drives set their own prices So legally they are not a ride share biz but a marketplace Founder is a lawyer who interned at the White House. He figured this out! P.s. this is NOT a paid promo. Just had a great experience that is all.
Andrew Yeung tweet media
English
152
129
5.3K
805K
0xCletus
0xCletus@leftcurvecletus·
@andruyeung I wonder what the trade-offs are on safety, insurance, support, etc
English
1
0
9
5.3K
Connor Burke
Connor Burke@cmb·
@UnslothAI Technically Lite here - how is this different than LM Studio? I see training locally which is new but interested in the substantial differences except for that
English
1
0
0
186
Unsloth AI
Unsloth AI@UnslothAI·
Introducing Unsloth Studio ✨ A new open-source web UI to train and run LLMs. • Run models locally on Mac, Windows, Linux • Train 500+ models 2x faster with 70% less VRAM • Supports GGUF, vision, audio, embedding models • Auto-create datasets from PDF, CSV, DOCX • Self-healing tool calling and code execution • Compare models side by side + export to GGUF GitHub: github.com/unslothai/unsl… Blog and Guide: unsloth.ai/docs/new/studio Available now on Hugging Face, NVIDIA, Docker and Colab.
English
216
833
5K
1.5M
Polymarket
Polymarket@Polymarket·
We're excited to announce 'The Situation Room' by Polymarket is coming to Washington, D.C. The world's first bar dedicated to monitoring the situation. 🧵
Polymarket tweet media
English
1.7K
2.3K
27.8K
33.9M
Connor Burke
Connor Burke@cmb·
@andruyeung Not the FlixBus! Rode it back Sunday due to Amtrak being sold out. Bathroom broke so they forced us to stop a few extra times… Used to have a bus called “The Jet” which was upscale with alcohol and gyro seats you would’ve loved
English
0
0
2
1K
Andrew Yeung
Andrew Yeung@andruyeung·
I was supposed to land at Newark last night but instead they kept our plane in a holding pattern for an hour before randomly diverting us to Washington DC at 1:33 am All while flying with through 54 mph winds with surreal turbulence. It was an exhilarating experience Apparently 22 other flights diverted to DC that early morning and every single hotel was sold out and train tickets to NYC were $500 They also lost my both my suitcases which added to the thrill Now I’m on a cramped 6 hour flix bus to New York. In total ~20 hours of commute time But this feels like a rite of passage of getting to New York City. It can’t be too easy; you have to work for it Because if you can make it to NYC you can make it anywhere I hope they find my luggage soon.
Andrew Yeung tweet media
English
82
19
862
293.4K
Connor Burke
Connor Burke@cmb·
@tech_savvy_guy_ Depends if the domain is rare. I’d recommend building since if u give up building or realize it’s not feasible than you’re stuck with a domain. That being said domains are cheap so up to you!
English
1
0
1
29
Soham Datta
Soham Datta@tech_savvy_guy_·
i have a cool idea for an app! should i buy the domain first or start building first?
English
1
0
1
191
Soham Datta
Soham Datta@tech_savvy_guy_·
hey @rauchg - there seems to be a flaw in the deployment protection in @vercel i am part of an organization in both GitHub and Vercel with my organization email added to both platforms. Now, when I author commits to my repo with my organization email, it blocks the deployment
Soham Datta tweet media
English
2
0
12
11.4K
Archie Sengupta
Archie Sengupta@archiexzzz·
Introducing AutoVoiceEvals I've applied the @karpathy autoresearch loop to voice AI agents. It's open source. Your voice agent has a system prompt. That prompt determines how it handles every call - bookings, complaints, edge cases, background noises, long pauses, people trying to trick it. Most teams write it once, test manually, and hope for the best. autovoiceevals makes it a loop. One artifact (system prompt), one metric (adversarial eval score), keep what improves it, revert what doesn't. Run it overnight. Wake up to a better agent. > How it works: You describe your agent in a config file - what it does, its services, policies, and what it should never do. You don't write test cases. You don't define attack vectors. provider: vapi / smallest ai assistant: id: "your-agent-id" description: | Voice receptionist for a hair salon. Maria does coloring only. Jessica does cuts only. $25 cancellation fee under 24 hours notice. Cannot advise on skin conditions. Closed Sundays. From that description alone, Claude generates adversarial caller personas - each with an attack strategy, a voice profile (accents, background noise, mumblers, interrupters), a multi-turn caller script, and pass/fail evaluation criteria. The eval suite is generated once and held fixed for the entire run, like a validation set. > The loop: 1. Read the agent's current prompt from the platform 2. Generate adversarial eval suite from your description 3. Run baseline 4. Claude proposes ONE surgical change to the prompt 5. Push the modified prompt to the agent via API 6. Run all scenarios against the updated agent 7. Score improved? Keep. Same score but shorter prompt? Keep. Otherwise revert. 8. Go to 4. Run until Ctrl+C. The system sees its own experiment history. When a change fails, the next proposal knows what was tried and why it didn't work. We ran 20 experiments on a live Vapi dental scheduling agent. 0 human intervention. > Score: 0.728 → 0.969 (+33%) > CSAT: 45 → 84 > Pass rate: 25% → 100% > 9 kept, 10 discarded > Prompt: 1191 → 1139 chars (better AND shorter) You describe your agent. It figures out how to break it.
Archie Sengupta tweet media
English
62
78
1.1K
229.4K
Abdul Qadir
Abdul Qadir@qadir5000·
12 hours ago, around 1am, I found out about Paperclip. 12 hours later, I skipped sleep because I couldn't goto bed with all the things it's unlocking for me and my businesses. Thank you @dotta
dotta@dotta

We just open-sourced Paperclip: the orchestration layer for zero-human companies It's everything you need to run an autonomous business: org charts, goal alignment, task ownership, budgets, agent templates Just run `npx paperclipai onboard` github.com/paperclipai/pa… More 👇

English
70
200
3.4K
931.5K