Alejandro Crosa

One of the finest videos on YouTube that reflects how everybody, including the government, administration, courts, and NGOs, is engaging in corruption. https://lnkd.in/eRGUF93s

5

We just baked a new Neural Search Talks episode with Aamir Shakir, co-founder at mixedbread.ai! We discussed the story behind the company's name, the recipes for their retrieval models which have recently made a splash on the MTEB benchmark, and the future of multimodality and multilingualism in search engines. (link below). Especially relevant if you care about small, efficient embedding models, based on high quality training data, and if you are interested in binarization and Matryoshka-ization of embeddings for memory and storage efficiency. Check the recording on YouTube, and let us know who you'd want to see in our podcast series next. And also let us know if you'd like to see these Mixedbread embeddings working for you in the Zeta Alpha platform.

61

2 Comments

Struggling to measure the impact of your new feature? Watch now: https://lnkd.in/g8cbrn5W In our latest video featuring Arpit Agarwal, Ex-VP, Analytics at Khatabook, where he dives deep into the world of feature success measurement! HelloMeets - Upskilling platform for PMs. #productmanagement #featureanalytics #khatabook #HelloMeets

11

3 Comments

There have been more reports and hand wringing lately of people getting serious sticker shock for "serverless" workloads. People have been attracted to the "serverless" dream for quite some time for multiple reasons including both cost && devex. One is the exact opposite of the aforementioned pain -- that is an expected lower bill, however, for anything that is not a toy you will *never* realize these financial gains. Keep in mind many of these serverless providers are built *on top* of the big 3 clouds to begin with. Right now you can get an always-on instance for ~$4/month at most cloud providers which will scale infinitely better than serverless workloads ever will. This is even before you take into account that most most of these clouds have crazy good free tiers to begin with. The second part of this equation is "no servers/no devops". You don't have to use hosted "serverless" offerings to get the same developer experience. You just need an operating system that was built for the cloud. Unikernels offer a 'no ops', 'no servers' approach that remove all the time cost and foot-guns of provisioning servers. They also remove all the security/performance woes that come with using something like containers. Half of these serverless vendors don't even segregate traffic amongst their customers - you can totally just talk to other tenants instances (which is crazy!!) (insert "We Take Security Very Seriously"™ here). Then there is the idea of cloud/vendor lock-in. I can take the exact same app and deploy it to 13 different providers in seconds with unikernels. This is where the hypercloud idea really starts kicking in with unikernels. With unikernels - instead of provisioning servers you just provision apps. While you can configure it to your hearts content you are not managing them and that is a massive difference. You don't have weird limitations like not being able to use ipv6 nor do you have crazy arbitrary limits on requests/second or request timeouts or any nonsense like that. Then there is the idea of "persistence" which most of these serverless offerings don't do out of the box - the cloud-native buzzword which is a euphemism for being able to do something super basic and necessary like run a database (such a shocker that you would need to attach a database to your webserver). Do yourself a favor next and instead of pushing your app out via serverless - start with https://ops.city unikernels instead.

24

Created an AI assistant app to connect patients and doctors using Vercel AI SDK, Assistant-UI.com components, function calling, thread suggestions, check it out (Guillermo Rauch, Simon Farshid, Kajari Ghosh)

10

A super clear, straight forward example of how to do cost effective RAG using LangChain for your document pipeline and Groq for compute resources. The code demonstrates how to speed up RAG and reduce costs using quantized embeddings (reducing 32 bit values to 8) Nice job Sahibpreet Singh! https://lnkd.in/gf5QMCKK

44

2 Comments

Big launches and loyal communities aren’t just for Apples, Microsofts, and Googles of the world. It’s for MicroSaaS, TOO. Cal.com, Inc. is a fantastic example of this. Dream big, and emulate the boldness of the FAANGs. 1) Build a loyal community from the start Peer Richelsen built Calendso, an open-source scheduling infrastructure, to address gaps left by solutions like Calendly, such as limited control and integration capabilities. They started with an early access waitlist on Slack, which turned into a community of enthusiastic evangelists, often sounding boards, and later their biggest cheerleaders during the launch. Start with a small group of engaged users who can give feedback and build excitement for your launch. 2) Plan and execute big launches Calendso launched its first product on Product Hunt in 2021, ranking it as Product of the Day, Week, and Month, establishing its presence in the market. But that’s not the most interesting part… 3) Listen and adapt fast After their launch, they listened to their customers and understood that… …sharing scheduling links necessitated a shorter domain name, which ultimately became their brand name—CAL! → Finding a three-letter domain in sync with their purpose was almost impossible, but it happened. Stay flexible and responsive, and don’t hesitate to make significant changes if they better align with users. 4) Rinse and Repeat - Validate continuously. Since its debut, Cal, has had 6 BIG WELL PREPARED launches despite its soaring success, including its latest scheduling marketplace kit with Supabase. Each launch has kept the company's community engaged and its product relevant. Successful MicroSaaS products aren’t static. Treat every version, feature, and update as a mini-launch. Iterate, validate and keep your audience engaged. Since then, they’ve transitioned from MicroSaaS and bagged a Series A of $25 Million in 2022. So, don’t hold back. Think big, launch boldly, and build strong communities. This is the only way to keep your product thriving. Are you launching your features boldly enough? #MicroSaaS #ProductLaunch #SaaS #CommunityBuilding #StartupGrowth

13

1 Comment

Proud that Codeflash got a shoutout in this awesome article by Harpinder Singh and Dhruv Iyer on AI-powered software engineering tools 👇 They dove into how AI tools like ours are effortlessly transforming the way developers work, making coding faster and more efficient -- and they highlighted Codeflash for helping engineers continuously optimize new code in every PR they open. Check it out! ⚡️ https://lnkd.in/gmuqsPjx

11

5 Comments

Exciting times for startups in the tech industry! Microsoft's latest announcement could be a game-changer for those who are flexible and agile enough to keep up with the evolving technology stacks. While it may seem like these advancements are making developers' jobs easier, it's important to remember that tech teams now require a diverse set of skills, including resilience and mental fitness, to manage the pace of constant change. Microsoft 's latest developments at the link below! 😎💻 https://lnkd.in/e_nXwk4K

5

1 Comment

Ever wonder why some companies can ship software at lightning speeds while others seem stuck in the mud? 🚀⏳ . . . . . . . . . Daksh Gupta of Greptile collected real-world insights from engineers at many of the world's largest tech companies, unpacking the intricacies that either propel or impede the speed of shipping code. The big reveal? Older and larger companies tend to move more slowly because the costs of missteps become greater. In some cases, those costs are fairly non-negotiable (e.g. ensuring software for customers in production remains secure) while in other cases, the costs are more optional and cultural (e.g. wanting a lot of accurate estimates, forcing a manager's vs. maker's schedule, etc). #SoftwareEngineering #Tech #DevelopmentVelocity #Innovation #Productivity

1

Jai OpenAI!! 🙏 Insanely Insane is here. GPT-4omni is out💯 I believe OpenAI has taken Tom Cruise lines literally to their heart: "I feel the need... the need for speed." Some crazy numbers they achieved in multilingual, audio and vision capabilities: 😎 1. 88.7% 0-shot COT Massive Multitask Language Understanding, like its around 8.28% higher than Gemini Pro 1.5. 2. 94.2% AI2D in test, compared to 80.3% of Google Gemini Pro 1.5 and 88.1% of Anthropic Claude Opus in Vision understanding. 3. Audio ASR performed much much better and effective, resulting over 100+% in Sub-Saharan Africa and around 55% increase in South-Asian audio. 🔴 IMP Noteups: 1. GPT-4o’s text and image capabilities have started to roll out yesterday in ChatGPT. 2. They are rolling out a new version of Voice Mode with GPT-4o in alpha within ChatGPT Plus in the coming weeks. 3. Developers can access GPT-4o in the API as a text and vision model. Looking forward to learn and delve deep into the model! 🤐 VC - OpenAI #openai #gpt4o #gpt4T #sora #ai #developer #software #ainews #northeastern #google #meta #softwaredeveloper #startups #investment #mlops #datascience #technews #data #api #perplexityai #harvard #california

31

It is important to revisit concepts that might have eluded me. Take the Intuit Lacerte Software Development Kit (SDK). For the non-programmers, a Software Development Kit is instructions and tools for connecting to local data sources. Think of them as APIs but for accessing local data and hardware instead of cloud data. The first time I looked at the Lacerte SDK over a decade ago, I did not understand how it worked. The best I was able to do was to pull basic client information (names and address) for generating mailing labels. This was before my firm was doing digital delivery. As I started using Robot Process Automation (RPA) tools to input data into Lacerte, I learned about batch input. This let me gain a new understanding of the structure of Lacerte's data. Just last winter, I built some RPA tools to read organizer PDFs from Lacerte to build Document Requests List in SafeSend and Firm360. This let me gain a new understanding of the flow of Lacerte's data. Just a month ago, I connected with TruePrep AI to build tools for inputting data into Lacerte. My original plan was an RPA method since it was something I was familiar with. In that moment, I had decided to take another look at the Lacerte SDK. With an additional 10+ years of experience in understanding Lacerte's data structure and data flow, the SDK all makes sense. In my excitement of applying new knowledge, I can now generate those previous Document Request Lists more efficiently as well as add, update, and delete data from Lacerte. What does this mean? It means, custom import tools that can be catered to a firm's workflow like trial balances, depreciation, etc. It is always important to revisit concepts that have been previously elusive. We are the sum of our experiences and even a few months (from winter to spring) can change how we find solutions to problems.

31

3 Comments

LLM, gen ai and copilot and all is fine, but a great reminder this morning on our team slack..of what REALLY matters, in this pic shared by our founding engineer Farhan Kaskar! specially when you are building a company which will help users make decisions on large volumes of data, interactively in real-time....you have to go back to the basics of good infra and optimized code..llm or no llm. and no better person than Farhan Kaskar to drive that discipline.

37

2 Comments

A healthy planet and healthy people are two sides of the same coin. – Dr. Margaret Chan As the summer solstice nears, we're reflecting on the healthtech advancements of 2024, and so far, it's been el fuego! Still though, the team is ACTIVELY preparing for what's next! It's inspiring to see how these innovations are reshaping global healthcare, transcending borders and seasons. 💡 Have you ever wondered how the relationship between day length and temperature relates to health science? Just like how our days warm up and lengthen, healthtech is expanding its reach and impact. We constantly think about the World Health Organization data about the millions of lives digital health could save annually. This is the revolution we're part of. 💡 Could seasonal shifts inspire breakthrough technologies in health? When temperatures rise and fall, so do the needs and solutions in healthcare. We've examined and pre-engineered concepts like ML/AI diagnosing with super accuracy and health prediction weartech. These sorts of concepts represent lives saved. 💡 Why should healthtech be seasonal? Our mission is to make sure that no matter where someone lives, they're covered. This means designing solutions for diverse climates and conditions, just like nature adjusts to the solstice. The power of healthtech lies in its universality. Another study found that telehealth is going to grow SO much in the upcoming seasons. This is all a testament to how quickly we can adapt and innovate when needed. So we urge you, let’s not wait. Let’s connect and create proactive, universal solutions. We believe our next breakthroughs will be born from understanding these natural cycles. Where health solutions are as dynamic as the seasons. Thank you, thank you for being part of this journey. 🙏 Ready to transform healthtech? Join us. Looking forward to continuing to innovate, build, and improve global healthcare together. 💛 📢 Afterthought: When you empower health, you empower the world and transform lives. #healthcare #healthtech #digitalmedicine #digitalhealth #technology and #innovation for #GlobalWellness #TechRx #Q2Flow The Equinox and Solstices Healthtech Connection May 2024

2

How India’s digital public infrastructure is enabling the emergence of state-backed ‘alt big tech’ "India Stack" Much of what was described as digital public infrastructure in this context flowed from India’s experiences with projects like Aadhaar for digital identification, the Unified Payments Interface for digital payments, and more recently, the account aggregator framework for data management in the financial sector. A host of such interventions built around the idea of standardisation of “open” application programming interfaces, or APIs, are collectively described as the “India Stack”. #digitalindia #upi #DPI #bigtech https://lnkd.in/g7tqyjSe

8

(Part 01) up, watch it in your free time, Big shout out to Whitney Lee, Kat Morgan, David Stenglein Marino Wijay, and Vamshi Krishna samudrala for sharing your knowledge and professional experiences, and lessons learned from notable industry events. Platform Engineering (unification of people, processes, and platforms) is a curated experience for engineers (consumers of the platform). Platform engineering starts with the people (the perceptible layer that hides the complexity and reduces the cognitive load from people solving the business challenges for their customers) and it's a promise that Organizations (can create standardized workflows), optimize resource usage, and lower operational costs by creating a unified platform. It is crucial to treat internal (users) as the platform’s customers, continuously seeking their feedback and adopting a structured methodology. Start by investing time in deeply exploring and comprehending the organization’s technological environment. Identify existing pain points and collect requirements from internal users. The job of Platform engineers on the team is to deliver shared infrastructure, runtimes, and other services consumed by developers across the organization with a shared responsibility model incorporating an accessibility matrix.  With insights from internal users, develop a comprehensive roadmap for the platform, outlining clear milestones and defining success criteria for each milestone. Platform(s) initiative dictates (product mindset), which possibly means cross-functional skills (product, design, engineering), In many cases team structure determines outcomes (rather than output). Platform Engineering is not a one-size-fits-all concept. It's about improving automation within a company and optimizing processes. Every company has its unique characteristics and challenges, and platform engineering aims to address these by establishing common standards and leveraging shared solutions (decisions should be based on your business requirements and priorities).   Platform engineering initiative dictates the importance of a product thinking mindset, Product thinking is the journey from the problem space of the users to the solution space of the business. Sometimes we need a trade-off: - Listen to customers (for user problems) - Don't listen to customers (for solutions) Watch the talk: https://lnkd.in/eqrBF9en #platformorchestrator #platformascode #platformengineering #backstage #infrastructureautomation #infraascode #policyascode #internaldeveloperplatform #developerportals

22

3 Comments