Black in AI

Learn and Network Previous volunteers learned academic and networking skills essential to their career goals!

Entrepreneurs move humanity forward by daring to do the scary, risky and crazy things. It was great rubbing minds with these folks at the AIX Ventures founders summit. Martin Weiss , Nasim Rahaman , Mohamed Hassan Kane, Nancy Xu , Pieter Abbeel

How might software engineering change with advances in AI that can write code? Will AI replace software engineers? I get versions of this question quite often and have gathered my thoughts in a blog post (WIP). TLDR; It covers three areas : 1. What AI can do today? AI excels at writing functions, refactoring code, and fixing minor bugs, but struggles with complex systems and niche problems. 2. What will change? The field will shift towards AI-assisted development (and agents via tools like #autogen), with more focus on code verification, model benchmarking, red teaming, design docs for machines, technology choices driven by models, model/data marketplaces etc 3. How should individuals and teams adapt? Engineers should enhance their critical thinking, and communication skills, , utilize models as tools, focus on the user, stay creative while teams need to develop frameworks for AI integration and invest in benchmarking/redteaming tools. References [1] Multi-Agent Systems with Autogen https://lnkd.in/gGSWsdym [2] Will Generative AI Replace Software Engineers? https://lnkd.in/e5-rvMPJ #multiagentbook #generativeai #softwareengineering #codignassistants

South African 🇿🇦 data scientist Pelonomi Moiloa has launched a groundbreaking large language model (LLM) designed to make AI more accessible and relevant to Africans. Moiloa is the CEO and co-founder of Lelapa AI, a Johannesburg-based startup building artificial intelligence tools for native African languages. Lelapa AI has just introduced InkubaLM, an LLM designed for five African languages: Swahili, Yoruba, isiXhosa, Hausa, and isiZulu. Working alongside her technical co-founder Jade Abbott, the duo has made significant strides in creating a platform that serves a combined audience of 364 million speakers. Read more on the UrbanGeekz story ➡️ https://lnkd.in/dtFpW4VK #artificialintelligence #ai #Africa #LanguageTechnology

It was awesome building Sauce with Colin Webb, and he tells our story (and shares our gratitude) very well. I strongly encourage you to check it out on his page! As mentioned in my previous post, the experience was both rewarding and incredibly educational. I've provided some of my own learnings as well here:

WARNING: High levels of grit, growth and gratitude! Earlier this week we announced that Sauce had been acquired. Today I’m sharing my story of how Sauce came to be, along with the lessons I’ve learned in this journey. From nights sleeping in the office, to eating PB&J and Soylent daily, to meeting our first investor on an airplane, to building a team and scaling nationwide with a solution that impacted millions of annual delivery orders, to getting acquired—I’m excited to finally share moments along this journey that I haven’t shared before. Most importantly, I was very lucky to be in the presence of so many great people who each helped shape Sauce, along with my personal life in the last few years. I couldn’t thank you enough! Thank you to my cofounder Nenye Anagbogu who’s been with me since we were dreaming of moments like this back in college. Thank you to my amazing early teammates Aaron Huang, Jason Trieu, Cindy Michelle, Vincent Wendy, along with Ian O'Neil, Wiwi Clarissa, Steven tan, Melissa Glander who helped us build something truly great. Thanks to our earliest believers Kashish Gupta, Jacob Becraft, Clarence Tatum, David Rosenthal, Riley Newman and more; our incredible venture investors Harlem Capital with Henri Pierre-Jacques and Gabby Cazeau, Red Sea Ventures with Scott Birnbaum and Paul Strachman, Rackhouse Venture Capital with Kevin Novak, Global Founders Capital with Roel Janssen and Don Stalter, and Ore Adeyemi; our expert advisors Lowell Smith, Ellie Doty, Juan George, Saul Cooperstein, Josh Seidenfeld; our early customers and partners Debbie Roxarzade, Kent Asaki, Jason Profitt, Sterling Douglass, Noah Glass, Sean J. Marrero, and more; and now Robert Earl and the VDC team—I’ve learned so much from each of you. Thank you for trusting us while our ideas were young! This has been quite the experience. Check out my personal story of how Sauce came to be: https://lnkd.in/gGUmrBeY

It’s been a thrilling few years building Sauce, and today we announced a big update. We’ve been acquired! I’m grateful to have worked with my amazing team to bring something new and necessary to a market that’s seen so much change. Over the last few years, every restaurateur was challenged to become an e-commerce company—something that requires a lot of technical expertise that wasn’t easily accessible within the industry. Fortunately, we built a solution to bridge that gap and help hard-working restaurant owners make the most out of their data and completely transform their online businesses. The new Linked Eats is truly an example of being Better Together, and it really is just the beginning. Our entire team is joining Linked Eats, and we’ve already reached an unprecedented scale after just a short period of working together. Stay tuned for new announcements soon! As for me, this was my first time building a company from scratch, and boy it was a hell of a ride. I’ll be sharing more about my founder journey within the next few days (and thanking the many people who helped us get here), but in the meantime, check out the new Linked Eats below! 

Forbes cover this month! Look at God 🙌 I feel incredibly grateful for the past two years and excited for what's to come. Our mission is nothing short of small, and we're committed to driving the positive change we want to see in Africa and the Global South. Thank you to all our investors, partners, and supporters; Amini is only getting started, and we have some exciting news to share with the world very soon 🚀 🌍 https://lnkd.in/dUChCnP6 #Forbes #Cover #ForbesAfrica #Grateful #AI #Africa

Intron Health, a Nigeria AI company that provides speech-to-text transcription tools for healthcare workers, has raised $1.6 million in a pre-seed funding round. The startup will use the capital injection to hire more employees, deepen its research efforts, and strengthen its cloud-native and on-prem capabilities.

In this new article, I and co-authors Boris Hanin, Lingjiao Chen, Peter Bailis, Ion Stoica, and Matei Zaharia explore one of the most powerful ideas we have yet discovered to inform compound AI systems design: verifiability. In common situations where practitioners are willing to expend a higher budget to go beyond the capabilities frontier accessible to today's state-of-the-art (SOTA) monolithic models, they may be willing to invoke many model inference calls, composing them into “networks of networks” (NoNs) of sorts. The question then becomes: what principles should guide the composition of these NoNs? Inspired by TCS and PCP notions that often verification is easier than generation (as holds for classical problems like graph coloring), we construct “best-of-K” or “judge-based” Compound AI Systems, which explicitly separate “generator” modules from “verifier” modules. We posit that these systems are particularly helpful for “reasoning-based” or “procedural-knowledge” oriented tasks, which are often more verifiable, less so for factual or declarative-knowledge settings (and we can use these systems partly to help characterize tasks, including subjects in the MMLU, along these lines). Very neatly, it turns out we can analytically characterize when these systems can confer a gain and predict the gain’s extent. We hope people will extend these ideas to tackle some of the reasoning-oriented application frontiers that are a bit beyond the range of today’s SOTA models. https://lnkd.in/gt5BbD4X