Ahmed Aly

9 reasons why Rana shouldn't have been a success. Being born by a mother, one of the first programmers in the Middle East is a kicka** start to a story. And Rana surely continued in that spirit. gpt5 >>> ❌ Non-commercial academic founder background   ❌ Female CEO in deep-tech AI (underrepresented at the time)   ❌ Emerging, undefined market category (“Emotion AI”)   ❌ No clear early product–market fit or vertical focus   ❌ Hardware-dependent early prototypes (glasses, sensors) increasing R&D cost   ❌ Long enterprise sales cycles (especially automotive)   ❌ Ethical / privacy concerns tied to emotion detection   ❌ Strong researcher identity — perceived as “too academic” for venture scale   ❌ Mission-driven vs. profit-driven narrative, creating investor skepticism 𝗥𝗮𝗻𝗮 𝗲𝗹 𝗞𝗮𝗹𝗶𝗼𝘂𝗯𝘆 𝟭𝟵𝟳𝟴 — Born in Cairo, Egypt.   Her mother was one of the first female computer programmers in the Middle East. 𝗘𝗮𝗿𝗹𝘆 𝟮𝟬𝟬𝟬𝘀 — Completes PhD at Cambridge. Joins MIT Media Lab’s Affective Computing Group, helps build early prototypes like “𝗲𝗺𝗼𝘁𝗶𝗼𝗻𝗮𝗹 𝗵𝗲𝗮𝗿𝗶𝗻𝗴 𝗮𝗶𝗱𝘀.” 𝟮𝟬𝟬𝟵 🦄 — Co-founds 𝗔𝗳𝗳𝗲𝗰𝘁𝗶𝘃𝗮 with 𝗥𝗼𝘀𝗮𝗹𝗶𝗻𝗱 𝗣𝗶𝗰𝗮𝗿𝗱 (MIT spin-out).   💰 Series A ≈ $2 M (Horizons Ventures, Kleiner Perkins, WPP, Fenox).   ⚡ Challenge: converting research into a commercial product and narrowing the vision.   🗣️ “The lab director said, ‘No, no, no. This is not research anymore. You’ve got to leave the lab, like start a company.’”   🗣️ “When you were first pitching Affectiva to investors they were often put off by your long list of potential use cases.” 𝟮𝟬𝟭𝟬–𝟮𝟬𝟭𝟲 — Growth and focus era.   💰 Series B $5.7 M (2011), Series C $12 M (2012), Series D $14 M (2016).   ⚡ Challenge: learning to focus on one solvable problem instead of chasing every opportunity.   🗣️ “We entertained all of them, but we quickly found that that was a recipe for disaster … you do have to focus and solve a specific problem for somebody.” 𝟮𝟬𝟭𝟲 — Becomes 𝗖𝗘𝗢 of Affectiva.   💰 Total capital raised ≈ $33 M.   ⚡ Challenge: leading a growing team and building operational maturity. 𝟮𝟬𝟭𝟵 — Raises $𝟮𝟲 𝗠 led by Aptiv, bringing total to ≈ $53 M.   ⚡ Challenge: entering the automotive industry with long certification cycles and conservative partners. 𝟮𝟬𝟮𝟭 🦄 — 𝗔𝗳𝗳𝗲𝗰𝘁𝗶𝘃𝗮 𝗮𝗰𝗾𝘂𝗶𝗿𝗲𝗱 𝗯𝘆 𝗦𝗺𝗮𝗿𝘁 𝗘𝘆𝗲 (𝗦𝘄𝗲𝗱𝗲𝗻) for $𝟳𝟯.𝟱 𝗠.   💰 Annual revenue ≈ $5 M with 86 % gross margin.   ⚡ Challenge: reconciling pride, exhaustion, and guilt around selling the company.   🗣️ “You have feelings of guilt — like you’re giving up on your team and your vision … I underestimated the amount of energy that would go into it.”  𝗣𝗼𝘀𝘁-𝟮𝟬𝟮𝟭 — 𝗘𝘅𝗲𝗰𝘂𝘁𝗶𝘃𝗲 𝗙𝗲𝗹𝗹𝗼𝘄 𝗮𝘁 𝗛𝗮𝗿𝘃𝗮𝗿𝗱 𝗕𝘂𝘀𝗶𝗻𝗲𝘀𝘀 𝗦𝗰𝗵𝗼𝗼𝗹, and 𝗲𝗮𝗿𝗹𝘆-𝘀𝘁𝗮𝗴𝗲 𝗔𝗜 𝗶𝗻𝘃𝗲𝘀𝘁𝗼𝗿.   ⚡ Challenge: redefining identity beyond the founder role.   🗣️ “My identity was so wrapped in Affectiva … I’m going through an identity crisis.”

8

For everyone in Lebanon asking me about freelancing Here’s a quick brief from my experience — with a promise to share more detailed insights once I have time for content. My start I self-learned programming, applied to local jobs on LinkedIn, got interviews, passed tests… but at the last step? Ghosted every time. No idea why. Maybe too old for a junior role. Maybe my lack of skills then, other reasons. So I went all-in on Upwork: studied tips, applied daily, and landed my first client in 3 days. Some people take 1–3 months — that’s fine. It works. I used the same approach for my brother; he got his first client in 2 months. After the first client It gets easier — but harder in new ways. Early on: low income, lots of hustle. Expect 6–12 months of grind before stability. Myths • “Freelancers work whenever they want.” Reality: To stay safe, you usually work slightly more than full-time. Keep one extra client above your comfort zone, because even “long-term” ones (months or years) can vanish. • “Freelancing is just small gigs.” Reality: No. It’s real product work in teams (designers, devs, PMs) — same as employment, just different contract terms. Contract types • Hourly — track time and get paid for every minute. 12h = 12h paid. 3h = 3h paid. • Milestone — fixed package of tasks, paid once delivered. • Weekly retainer — like booking your time. Example: 15h/week at $10/h = $150/week guaranteed, whether they give you full tasks or not. Extra hours billed on top. Why I prefer it over employment • Diversify risk: instead of 40h for one company, split across 2–3 clients. • More experience, tools, networks. • Multiple income streams + stronger portfolio. • Remote flexibility. • Fewer interview loops. Once proven, clients come to you. The catch • No paid vacations/sick days. • You must deliver the hours you commit. • Strong time/energy management. • Always in push mode — I worked Sat/Sun for 2 years; only recently I take Sundays off. But if you’re sick or need a day, you can shift your schedule. That flexibility is real. One more myth That “laptop lifestyle” — $100K from the beach, 2h/day — is marketing to sell courses. That’s not freelancing. It only appears when you build a business: a team, delegation, and stepping away from daily work. Freelancing is grind: coding, design, client calls. The upside: you control clients and schedule. I’m building toward the business side with my agency, but freelancing itself isn’t a shortcut. Employment or freelancing First? No single answer. Some say: get a job first, then freelance. Others (like me): start freelancing from day one — yes, even as a junior. How to start • Learn everything you can about freelancing & Upwork. • Apply consistently. • No views → fix proposals. • No interviews → fix profile/portfolio. • Interviews but no offers → fix sales pitch. • Sometimes it’s luck or competition — focus on what you control.

60

7 Comments

From Idea to Reality: Building an Arabic AI Voice Assistant After diving deeper into AI development, I realized something obvious but rarely addressed: Arabic voice AI is still far behind English. So I decided to build one myself. Introducing Mujīb AI : a production-ready Arabic voice assistant. It took a lot of trial-and-error, debugging, and late-night fixes, but it’s finally real. What it can do: Automatically answers phone calls in Arabic, English, Urdu, and French Books appointments in real time Provides instant knowledge retrieval (RAG) Includes a full analytics dashboard Tech Stack: Python (Whisper STT, LLaMA 3.1) • C# .NET Core • PostgreSQL • Twilio • RAG It’s not perfect yet, TTS still needs improvement, and response times can be faster. But it works, and building it taught me more than any course ever could. Question for the Saudi tech community: Anyone here has experience integrating with local telecom providers like stc, Zain KSA, or Mobily? I’m using Twilio for now, but I’d love to connect directly with local infrastructure. #AI #SaudiArabia #TechInnovation #VoiceAI #Arabic #رؤية_السعودية_2030 #MachineLearning #Startup

6

4 Comments

Introducing hikma-pr (Local AI assisted pull request reviewer): Hikma-PR is a sophisticated, stateful CLI application designed to automate the review of GitHub Pull Requests. Its core mission is to leverage the power of locally hosted Large Language Models (LLMs) to provide in-depth, multi-faceted code analysis without relying on cloud-based services. This gives you unlimited, cost-free, and private code analysis. GitHub: https://github[.]com/foyzulkarim/hikma-pr (https://lnkd.in/gT9p8m3S) NPM: https://www.npmjs[.]com/package/hikma-pr (https://lnkd.in/gFDq4CYq) Please let me know your thoughts. More updates coming soon. #hikmapr #hikmacli #lmstudio #llm #localllm #pr #pranalysis

65

1 Comment

Mastering .NET and C# shouldn't be complicated! 💻 Continuous learning is the key to staying ahead in the software industry. I’ve always found that the best way to learn is through practical, real-world examples. On my YouTube channel, SarrawyDev, I focus on breaking down complex topics into simple, actionable tutorials. From ASP.NET Core and Entity Framework to building mobile apps with .NET MAUI, I cover it all. 🚀 What you’ll find: Step-by-step CRUD operations. Integrating SQL Server with C#. Advanced tips for .NET developers. Check out the channel and let's level up our coding skills together: 🔗 https://lnkd.in/dtRtiyYA #DotNet #CSharp #AspNetCore #SoftwareEngineering #SarrawyDev #Programming

2

🚀 Just launched AI Code Reviewer API – V1 & V2! Powered by Ollama, it gives smart, versioned code analysis right from your API. 💡 Phase 1 highlights: Clean, versioned endpoints (Swagger-ready) Modular design with DI Sample code review in action Phase 2 is coming with more AI-powered enhancements! Check it out 👉 https://lnkd.in/e2zgedbc #DotNet #AI #CodeReview #Ollama #DeveloperTools #API #SoftwareEngineering

1

Does nature's most famous constant hide inside the structure of elliptic curves? 🌻 I am thrilled to announce the publication of my final research, "Brahim's Theorem: Golden Ratio Scaling in Elliptic Curve Arithmetic," now available on Zenodo. This theorem suggests that the mathematical universe of elliptic curves is more structured than previously thought. In cryptography, structure can sometimes mean vulnerability (if an attacker exploits it) or assurance (if we use it to prove robustness). Key findings from the paper: The Golden Ratio Connection: The probability of non-trivial $Sha$ groups follows a power law determined by $\phi$, with a scaling exponent of $\beta \approx 0.2406$. Data vs. Theory: Empirical evidence strongly favors this Golden Ratio hypothesis over the traditional binary entropy models. A New Model: The results suggest that attempting to map curve invariants to fluid turbulence (Reynolds numbers) is less accurate than the new Arithmetic Density Model proposed in this work. This research bridges the gap between abstract number theory and large-scale data analysis. It pushes us to ask: Is arithmetic geometry more "organic" than we thought? 📄 Read the full paper and access the datasets/code here: https://lnkd.in/dGjNQxew

3

1 Comment

In this new Engineering Blog, MongoDB Staff Research Engineer Will Schultz and Principal Research Scientist Murat Demirbas dig into MongoDB's support for distributed, multi-document transactions. They also discuss recent work on formal modeling and verification to examine transactions protocol in more depth. MongoDB employs TLA+ specifications to verify its distributed transaction protocol, prioritizing both correctness and permissiveness. By mathematically modeling system interactions, engineers ensured strict snapshot isolation while minimizing unnecessary transaction aborts. This approach ensures that the database maintains rigorous data integrity without sacrificing high-throughput performance. The key takeaway for developers: assuredness. https://lnkd.in/deXrRD2S

4

Learning with an LLM on your side is so very powerful.. I always appreciated and wished for access to experienced, really talented, and knowledgeable engineers.. Now imagine you learn and have the ability to ask a polymath engineer any question you want, at any time you want, and they will never get tired of answering with infinite amount of patience.. That's a really big deal in terms of learning.. Just provide some context, ask your question, and learn from the answer. Still confused? Ask follow-up question/s? Still don't understand? Ask to generate some visual examples. Ask for diagrams. Still don't understand? Ask for references or discussions on the internet related to your questions.. But, remember, this also raises the bar in terms of what's expected from you and how fast you can learn.. As you do your best, and use your best LLMs/tools for learning, others do that too..

18