PromptLayer @promptlayer
The first platform for prompt engineering. Collaborate, manage, and evaluate prompts. https://t.co/wxIrb05J3z 🍰 promptlayer.com NYC Joined March 2022-
Tweets362
-
Followers3K
-
Following273
-
Likes528
Bringing back one of my favorite events that we've run for NY tech week. AI Demo night ✨✨ Lightning demos of real LLM apps by my favorite NY founders. lu.ma/mzhzxc4t
Bringing back one of my favorite events that we've run for NY tech week. AI Demo night ✨✨ Lightning demos of real LLM apps by my favorite NY founders. lu.ma/mzhzxc4t
👉 Are your AI prompts truly delivering? Join @imjaredz, Founder of @promptlayer, at the MLOps Conference for his talk on 'Iterative Strategies to Testing Prompts.' Learn sophisticated evaluation techniques to enhance prompt effectiveness.
Cool LLM startup gave me this idea today... and obviously the best time to ship quick new features is Friday evening Enjoy the new side-by-side Github-style diff viewer in @promptlayer 🍰👀🤯
🧩 Complex jailbreak prompts are the most effective Prompts that mix multiple jailbreak tricks tend to work best for getting around ChatGPT's rules. But if they're too complex, the AI might get confused.
最大的提示词工程平台 - PromptLayer 来自 @promptlayer 创办人 @imjaredz @pranav_kanchi 帮助构建真正的 AI 应用程序,包括强大的监控、提示词版本控制、成本分析以及评估提示词有效性的方法。使用一行代码即可码开始记录、组织和评估您的提示词。 特色功能: 1. 提示词登记 -…
🧵 Building your own personal ChatGPT (best practices) ChatGPT is actually a way simpler piece of software than you would expect, if we take @OpenAI apis as magic 🪄 We've learned a lot about prompt engineering best practices from our customers. Trying to share that knowledge
It's easy to be an expert when nobody knows what's going on @imjaredz 😉
My super duper unofficial model benchmarks have arrived! Evaluated using this prompt and a 25 handwritten test cases. 🥇 WINNER: GPT-4-Turbo #2: Mistral #3: Claude Sonnet and GPT-3.5-Turbo tied #4: Llama3 Obviously a small sample size so take it with a grain of 🧂
Another day, another gamechanging LLM 🚨🚨🚨 Snowflake just released Arctic, their own open-source LLM. Why you should care: 💸 Run inference directly on Snowflake instance. Push LLM costs to the customer. Huge. 🫡 Really good at coding and SQL generation. ⚖️ Relatively tiny.…
We just added support for Llama3 on PromptLayer. @AIatMeta 📈📈📈 You can run use the playground, run evals, and compare it against GPT-4. The top 20 open source LLMs are now on PromptLayer 🍰 Thanks to @CloudflareDev for making this so easy!
Exciting stuff... @promptlayer is mentioned in the new Enterprise GenAI @Gartner_inc research. Some novel things in here 🧵 of what else they discovered
Llama 3 means it's time to promote this blog post again Migrate prompts to open source ! blog.promptlayer.com/migrating-prom…
Llama 3 is out. You can't ignore open-source models anymore. It's supposedly better than Claude 3 Sonnet 🤯🤯 Huge
Llama 3 has been my focus since joining the Llama team last summer. Together, we've been tackling challenges across pre-training and human data, pre-training scaling, long context, post-training, and evaluations. It's been a rigorous yet thrilling journey: 🔹Our largest models…
Building ChatGPT from scratch, with best practices ChatGPT is a simpler app than you would expect. In this tutorial we will build our very own personal ChatGPT. ➡️ With a special emphasis on prompt engineering best practices! blog.promptlayer.com/building-chatg…
Prompt engineering is not dead 😂😂 An article recently made rounds talking about some new auto-prompting tools and the false rumors of the death of prompt engineering. Love @simonw's response (see screenshot). Let me take this time to share MY definition: Prompt engineering…
Cool article on the AI Stack by @MongoDB and @richmondalake Love the PromptLayer shout out! AI is collaborative. Couldn't agree more. mongodb.com/resources/basi…
Whoa, I just realized that raising a kid is basically 18 years of prompt engineering 🤯
🏷️ Use XML tags Claude pays special attention to XML tag structure. You can use XML tags to clearly mark different parts of your prompt, like indicating which text Claude should remove personal info from.
Diana Alokhina @dianalokada
115 Followers 820 FollowingTashi Dikki @TDikki47973
9 Followers 91 Followinghitesh (e/trsfrm) �.. @hitmadeit
113 Followers 274 Following a python developer cum electronics engineer, building @neuraleverage to help people learn genai. dm me for genai/rag/chatbot/evals/backend gigsjeremyfelix @felixHcat
55 Followers 263 Following He's like some sort of strange robotic techno-kitten.Prashanth Koripalli @prashanthko7
5 Followers 130 FollowingAIWorld @zhuyunyun_
40 Followers 104 Following Former Senior Research Scientist at Google AI, specializing in cutting-edge AI applications and consulting.francisco barrallo @fbarrallo
44 Followers 308 FollowingAmir Haouara @St_Thomas_H
1 Followers 13 Followingbadri narayana @btbrigade123
25 Followers 2K Following what ever you are what ever you do, always be good and kind!Arsen Apostolov @arsenapostolov
21 Followers 56 Followingar4499 @ar4499_
0 Followers 163 FollowingCourtenay @courtenay100
138 Followers 1K Followingupsandowns @upsandowns9548
12 Followers 93 Followingojbtjbs @israulg
14 Followers 128 FollowingWayne Painters 🌻 �.. @PaintersWayne
1K Followers 2K Following Let’s do this! Follow me in the fight against sedition, hypocrisy and anti-democracy in an already great country. Don’t give in to the magat gaslighting! 🌊Mark Kovarski @mkovarski
2K Followers 5K Following Responsible AI, Cloud, SaaS, Product 🤖 🫶🌐💡 | https://t.co/2vuiFosXlm 📪Sam熊哥哥 @Sam65181116
41 Followers 2K Followingshowli @showli_peng
21 Followers 108 Followinghua @jaKehua117
111 Followers 2K FollowingGeng Liming @CoderLim
11 Followers 213 Followingstan_think @stankekulyh
0 Followers 1K FollowingTiger @hemingway888
6 Followers 2K Following菜刀 @MrQuanlijie
0 Followers 1K Followinghi @i_ndream
0 Followers 223 FollowingAnunnaki @Goodspeed2011
46 Followers 544 Followinggllgoe king @gllgoe
28 Followers 779 Followingimesong @imesong2012
922 Followers 2K Following https://t.co/YhdEa1c9kM #TikTokShop Software Engineer #Android Developer #FrontEnd #FullStackFurqan @trulyfurqan
16 Followers 372 Followingzzzzzzoo00oo @zzzzzzoo00oo
245 Followers 4K Followingbilly ha @billyha18
480 Followers 4K FollowingFrenlyInCA @FrenlyInCA88
387 Followers 3K Following EM doc in a past life. Escaped the rat race. Now, just an anon crypto degen. Above all else, be kind.Peter of Hovasabi @AkinolaPeteru
3K Followers 3K Following Futurist Building @Hovasabi_HQ || 3x founder || 2006 @time person of the year |🧘♂️bringing ideas to life using tech #Edtech✊ Featured on CNNCrypto450 @Crypto_450
50 Followers 756 FollowingKatia Ameri @KatiaAmeri
11K Followers 1K Following LA Persian girl meets SF tech bro. Recovering founder. Partner @a16z | Previously @Stanford @PearVC | Launching #NYTechWeek 2024, @techweek_Aston Zhang @astonzhangAZ
5K Followers 92 Following Research Scientist at the #llama team of Meta Generative AI, designing and training large language models. Opinions are my own.Arthur Mensch @arthurmensch
40K Followers 874 Following Co-founder and CEO @MistralAI. Apply https://t.co/yHGRZAtjcxclem 🤗 @ClementDelangue
91K Followers 5K Following Co-founder & CEO @HuggingFace 🤗, the open and collaborative platform for AI buildersSeb (whop.com) @SebCuadros
2K Followers 2K Following Community @whopio, positivity and optimism enjoyer. Hiring 10x TS/Next Web App Devs & Product Designers always (only if youre cracked (respectfully) (in nyc))Ajay Bhargava @ajay_bhargava
363 Followers 1K Following 🧑🏽🔬 who speaks 🧑🏽💻 natively. I also 🏋️,🧳, 👨👩👧. Equal parts: 🇨🇦, 🇬🇧, 🇺🇸, 🇮🇳Mimansa Jaiswal @MimansaJ
1K Followers 3K Following MoTS @normativeai. Ex @UMichCSE, 2x @MetaAI, @allen_ai | Speech & NLP | Robustness, Data & Annotations, Evaluation & Interpretability in LLMsAri Weinstein @AriX
17K Followers 8K Following Co-founder @SoftwareAppsInc. Previously managed Shortcuts and SiriKit at Apple, and co-founded Workflow. @[email protected]Morgante @morgantepell
2K Followers 3K Following Living life heuristically / Solving technical debt @ https://t.co/9UtllLfyAr / Previously @GoogleCloudrabbit inc. @rabbit_hmi
85K Followers 1 Following rabbit brings the future of human-machine interface. order r1, your pocket companion, now.Company Ventures @Company
11K Followers 2K Following NY’s foremost venture community. We combine space, programming, expertise, and capital to accelerate success for exceptional founders and startups.Ilya Sutskever @ilyasut
370K Followers 2 Following towards a plurality of humanity loving AGIs @openaiJason Levin @iamjasonlevin
24K Followers 997 Following I help startups/VCs go viral. Read about my adventures and strategies here: https://t.co/UWaWrsgQMk. Growing @producthunt, @jamdotdev, and more.mathurah @mathurahravi
6K Followers 992 Following prototyping experiments, ideas, words, and investing. currently @rabbitholeathon, @waterloosyde. built things @netflix, @replit, @cowboyvcthenewboston @thenewboston_og
2K Followers 136 Followingjason liu @jxnlco
19K Followers 1K Following sabbatical @southpkcommons, angel investor?? prev @stitchfix @metanick bradford @n_s_bradford
569 Followers 463 Following co-founder/CTO @ellipsis_dev (YC W24) | building AI software engineersMike Taylor @hammer_mt
6K Followers 2K Following 🤖 Building AI products at https://t.co/ZfiXgyO2xb 📚 O'Reilly author on Prompt Engineering 🎓 350k students have taken my courses 📈 Built a 50 person growth agencyNomic AI @nomic_ai
14K Followers 50 Following Building explainable and accessible AI https://t.co/bbYqCdL8vQAndriyMulyar @andriy_mulyar
11K Followers 517 Following building tech that enables humans to interact with latent spaces 🗺️ founder / cto @ https://t.co/NbsLHLWfy8 prev. ML Ph.D. Student at NYU Courantiamrobotbear 👺 @iamrobotbear
4K Followers 5K Following Product Manager working on Generative AI & machine learning. Opinions are my own, not my employer's. RT !=endorsementAbuBakar Sohail @Abu_BakarSohail
99 Followers 158 FollowingBalaji @balajis
1.0M Followers 4K Following Immutable money, infinite frontier, eternal life. #BitcoinDavid Chie @davidjchie
29K Followers 3K Following Australian-born. LA-living. Early adopter, recruiter, investor.Logan Kilpatrick @OfficialLoganK
92K Followers 2K Following Lead product for @Google AI Studio and working on the Gemini API, helping developers build with AI, my views!LlamaIndex 🦙 @llama_index
63K Followers 24 Following The way to connect LLMs to your data. Github: https://t.co/HC19j7vMwc Docs: https://t.co/QInqg2zksh Discord: https://t.co/3ktq3zzYII https://t.co/UXeIlwvvbAEric Tu 🧢 @EricWasNumber2
135 Followers 175 Following Founding Engineer @RogoData | YouTube channel manager for @BwipoHarris Rothaermel @DeveloperHarris
7K Followers 677 Following human supremacist, hyper techno-optimist, 🇺🇸 chief AI whisperer @starlightlabsco ✨Catalyzing Innovation @CatalyzingInnov
53 Followers 3 Following Cross-sector global inspiration hub. Curated by Michelle Greenwald.Rez Khan @RezwanKh
219 Followers 443 Following VP Product, Design @AbridgeHQ and Venture Capital https://t.co/lkEJxlPOsbMarco Mascorro @Mascobot
10K Followers 2K Following Partner @a16z | Cofounder @Fellow_AI | AI & Robotics Engineer | prev @BMW research | @MIT 35 under 35 | Opinions my own.Anthropic @AnthropicAI
263K Followers 26 Following We're an AI safety and research company that builds reliable, interpretable, and steerable AI systems. Talk to our AI assistant Claude at https://t.co/aRbQ97uk4d.Danyil Blyschak @danblys
370 Followers 1K Following No solutions only tradeoffs. Stealth mode. Erdős number: 4. DMs open :)Saleh Hindi @SalehOfTomorrow
892 Followers 2K Following tryna do the most good, friend. ex @lyft, @bloomberg. Building something new. In my ✨habibi era ✨Gabriel Stengel @GabeStengel
541 Followers 968 Following founder @rogodata / NLIs before they were cool @PrincetonCS / always looking for pickup basketballYusuf Olokoba @OlokobaYusuf
442 Followers 369 Following I like to build time machines. Building @fxn_ai.Rachel Woods @rachel_l_woods
21K Followers 1K Following Ex-Facebook Research Data Scientist. Pushing the envelope of operating businesses with AI (AI Operations). #AIOpsMarc Andreessen 🇺�.. @pmarca
1.4M Followers 24K Following Techno-optimist. E/acc. Technology brother. Move Fast and Make Things. p(Doom) = 0; p(“1984”) = not 0.Arize AI @arizeai
2K Followers 78 Following Arize AI is an ML observability platform enabling ML practitioners to better detect and diagnose model issuesKasra Kyanzadeh @kasrak
2K Followers 673 Following Exploring ways AI can improve individual lives Prev early team at @Airtable, @Watershed👀 Big news! The official calendar of events for #NYTechWeek is now LIVE, and we are so excited to announce that this is officially our largest Tech Week EVER! - 500+ events hosted by VCs & startups 🤯 - June 3-9 2024 🗽 Want the complete list of events? Register here for…
Bringing back one of my favorite events that we've run for NY tech week. AI Demo night ✨✨ Lightning demos of real LLM apps by my favorite NY founders. lu.ma/mzhzxc4t
👀 Big news! The official calendar of events for #NYTechWeek is now LIVE, and we are so excited to announce that this is officially our largest Tech Week EVER! - 500+ events hosted by VCs & startups 🤯 - June 3-9 2024 🗽 Want the complete list of events? Register here for…
👉 Are your AI prompts truly delivering? Join @imjaredz, Founder of @promptlayer, at the MLOps Conference for his talk on 'Iterative Strategies to Testing Prompts.' Learn sophisticated evaluation techniques to enhance prompt effectiveness.
heya @promptlayer it's exams rn, but I'd really like to organize a prompt building/ prompt jacking micro-competition at my university I'd like you guys to be a part of it, you guys up ? I can dm if interested
Cool LLM startup gave me this idea today... and obviously the best time to ship quick new features is Friday evening Enjoy the new side-by-side Github-style diff viewer in @promptlayer 🍰👀🤯
🔒 ChatGPT's restriction strength varies by topic ChatGPT is stricter about filtering out some types of content than others. The strength of its safety measures depends on the topic.
🆚 GPT-4 is more resilient than GPT-3.5 GPT-4 is better at resisting jailbreak attempts than GPT-3.5, but people can still frequently trick both versions into saying things they shouldn't.
🔄 Jailbreak prompts constantly evolve Whenever ChatGPT's safety controls are updated, people find new ways to jailbreak it. It's like a never-ending game of cat and mouse between jailbreakers and the devs.
🧩 Complex jailbreak prompts are the most effective Prompts that mix multiple jailbreak tricks tend to work best for getting around ChatGPT's rules. But if they're too complex, the AI might get confused.
🎭 Pretending is the most common jailbreak technique Most jailbreak prompts work by making the AI play pretend. If ChatGPT thinks it's in a different situation, it might give answers it usually wouldn't.
What can we learn from ChatGPT jailbreaks? Found a research paper that studies all the jailbreaks of ChatGPT. Really interesting stuff... By studying via negativa (studying bad prompts) we can become better prompt engineers. Learnings in 🧵 arxiv.org/abs/2305.13860
最大的提示词工程平台 - PromptLayer 来自 @promptlayer 创办人 @imjaredz @pranav_kanchi 帮助构建真正的 AI 应用程序,包括强大的监控、提示词版本控制、成本分析以及评估提示词有效性的方法。使用一行代码即可码开始记录、组织和评估您的提示词。 特色功能: 1. 提示词登记 -…
🚦 Regression tests are your friend Ensure your AI assistant maintains performance and reliability by setting up regression tests to compare new prompt versions against historical outputs.
🚀 Be intentional with production deployments Select stable production versions of your prompts carefully, and use release labels to easily manage deployments.
🧩 Modular prompting with snippets Organize your prompts into reusable snippets to maintain consistency, allow non-technical stakeholders to make edits, and streamline the development process.
📝 Version control is key Keep track of prompt changes using version control tools like @promptlayer to easily identify and revert to successful iterations.
■ Treat the LLM as a Black Box Approach prompt engineering by writing in natural language and checking the output. Trial and error is the name of the game. Don't plan too much.
The main point I want to make ➡️ Faster iterations = Better prompts Your prompt engineering dev environment matters. Faster iteration cycles is how you get to better prompts. These are the best practices I'll describe here blog.promptlayer.com/building-chatg…
🧵 Building your own personal ChatGPT (best practices) ChatGPT is actually a way simpler piece of software than you would expect, if we take @OpenAI apis as magic 🪄 We've learned a lot about prompt engineering best practices from our customers. Trying to share that knowledge
It's easy to be an expert when nobody knows what's going on @imjaredz 😉