I want to begin this book by sharing my background and how I found my way into technology and software development. This will give you a sense of my journey with technology and how it connects to the software development life cycle (SDLC).
After passing the French Baccalaureate exam (which is equivalent to completing a Freshman year in the United States) in July 1978, I packed my bags with three of my high school friends and moved from Zahle, Lebanon, to Fort Worth, Texas, to attend an English-language school. Arabic is my native language, and I attended French schools, so English was my third language. Passing the TOEFL exam was a prerequisite to my acceptance at the university. That was November 1978, and Lebanon at the time was torn by civil war. I was seventeen years old with a $250 money order in my pocket, and I had no option but to succeed or die trying. In March 1979, I passed the TOEFL exam and was accepted as a sophomore into the Computer Engineering major in the Electrical Engineering Department at Northeastern University (NU) in Boston, Massachusetts. NU had and still has one of the best cooperative education (internship) programs in the United States. I wanted to practice what I learned to gain an edge in the workplace. I landed my first co-op job at Honeywell Information Systems in Brighton, MA, and started on January 7, 1980. After three months of internship, I was offered a full-time position on the graveyard shift from 11:00 p.m. to 7:00 a.m. Sunday through Thursday, with a full-ride scholarship for my undergraduate studies. So, I happily took it.
While at NU, I learned programming in BASIC, Fortran, and Macro Assembler on the DEC VAX-11/780 minicomputer, which had a whopping 32 MB of RAM (yes, that is Megabytes). We were using either hardcopy console terminals with dot-matrix printing or VT-100 monochrome dumb terminals, if we were lucky and early in line at the Computer Center to grab a terminal. Then, we would line up at one of two punched-card photoelectric readers to feed them our code at an amazing speed of 300 cards per minute. Those days remind me of the soup line episodes on Seinfeld. You’d better be nice to the Computer Center operators if you want your code to be loaded and executed without delay.
Things got exciting in the summer of 1981 when a co-worker at Honeywell bought an IBM PC Model 5150 for about $3,000. It had 32 KB of RAM, a 5.25-inch floppy drive, and a monochrome monitor running IBM PC-DOS 1.0. I experimented with IBM’s Macro Assembler and BASIC, but couldn’t use IBM Pascal since it needed at least 48 KB of RAM, which was too expensive for my friend. In 1982, another friend got a Commodore 64 with 64 KB of RAM. I often borrowed it to write BASIC and Macro Assembler programs, saving them to its 200 KB data cassette—100 KB per side —which felt like a huge upgrade back then. By 1983, after saving up, I bought my first transportable computer, the Eagle Spirit, for about $5,000. That’s about $17,000 in today’s money. Looking back, if I had invested that money in the S&P 500, it would be worth over $780,000 now. Still, it was thrilling to own a personal computer with a 4.77 MHz Intel 8088 processor, a 5.25-inch floppy drive, a 10 MB hard disk, a 9-inch green monitor, and MS-DOS. It felt like real computing power at my fingertips. But don’t be fooled—it was like lugging around a giant oscilloscope. My “transportable” PC weighed over 35 pounds, or about 16 kilograms, so carrying it was a workout in itself. The Eagle Spirit was definitely transportable, but you couldn’t put it on your lap.
After graduating in 1983, I got a job in Saudi Arabia and the United Arab Emirates with Mideast Data Systems, the distributor of Data General minicomputers in the Gulf Cooperation Council (GCC) countries. I spent five years selling, installing, and implementing solutions in the healthcare, retail, energy, and government sectors. I was customizing COBOL and MIIS/MAGIC (based on MUMPS) applications. Then, I spent a year on similar projects with the local Prime Computer distributor in Dubai before immigrating to Montréal, Québec, Canada, in 1990, with my wife and newborn child. In this early stage of my career, I worked across all phases of the SDLC, primarily using the Waterfall method, the standard methodology pioneered in the 1970s.
Now that I carbon-dated myself, let me fast-forward to March 1991, when I landed a job in Montréal working for Ashton-Tate, the company that developed dBASE, the first database management system for the PC. Being English- and French-speaking was a major help in landing this job. Four months into my tenure at Ashton-Tate, our company got acquired by Borland International, its much smaller competitor. In November 1993, I was transferred to the Canadian headquarters in Toronto, where my primary role was to support the implementation of databases (dBASE, Paradox, and InterBase) and spreadsheets (Quattro Pro) across Canada. During this period, my focus was on the downstream deployment and support stages of the SDLC. I traveled frequently to the U.S. headquarters in Scotts Valley, California, and participated in several bug hunts that preceded the release of Borland products to manufacturing. I made more than $65,000 finding bugs, so management decided it was more cost-effective to offer me a job at headquarters and move me there than to keep paying me for moonlighting as a bug hunter.
In June 1994, I moved to California, where I became the international product manager for dBASE and Quattro Pro, its flagship products. In the same year, Novell acquired Borland's Quattro Pro business, along with the product team and me. In late 1995, I joined Starfish Software, a startup founded by Borland’s founder, Philippe Kahn, where I led the development of Internet Sidekick, a personal organizer for contacts and calendars, TrueSync, a utility for synchronizing personal calendars and address books between a PC and mobile devices; EarthTime, a world time clock; and other calendaring and personal productivity solutions based on the new Java platform at the time. I cherish Philippe’s mentorship, as I learned from him the spirit of perpetual entrepreneurship: to always aim high and never settle for good enough. My focus during this stage of my career had shifted more upstream to ideation, design, and development, while working with the Internet Engineering Task Force (IETF) on the iCard/vCard and iCalendar/vCalendar specifications.
In March 1997, I attended a barbecue event at a friend’s house in Felton, California, where I ran into Marc Randolph. He was a colleague of mine at Borland, where he led the Sidekick product team. Marc shared his idea for a DVD rental service with me and asked if I wanted to join his startup. He was very excited about the launch of the DVD format a few days earlier, which Marc saw as an opportunity to ship lightweight DVDs to clients by mail instead of the costlier VHS tapes. I was lukewarm about relying on the U.S. Postal Service (USPS) for back-and-forth DVD delivery. Why would I bother waiting for a DVD by snail mail if I could just run down the street from my house and get the latest-released DVDs from the Blockbuster Video store much faster than the USPS could deliver them to my home? I did not see the eventual streaming capabilities and high-speed Internet connectivity because, at the time, we were dealing with 33.6 Kbps dial-up modems, 56 Kbps if we were lucky, and 128 Kbps ISDN lines at the office! So, with my short-sightedness, I missed out on a great opportunity to own founder shares in Marc’s startup, Netflix. Marc eventually became a millionaire, and I became just me. Instead, in April 1997, I joined Agile Software as vice president of marketing, a job having little to do with the SDLC, except in the requirements phase.
In December 1998, while on Christmas vacation and at a family reunion in Ho-Ho-Kus, New Jersey, I got the flu. My wife and kids went on a day trip to New York City with her sister and her family, where they visited Ellis Island and Rockefeller Center. I asked them to take pictures with their digital cameras wherever they went. As I sat alone, sneezing and coughing, feeling miserable about missing out on the fun, I said to myself, “Wouldn’t it be nice if they could log in to a secure site, upload their photos as they took them, and share them with me so I could see them!?!?” Philippe Kahn had done something similar a year earlier when he took a photo of his newborn daughter Sophie using a digital camera, connected it to his Motorola flip phone, uploaded it to a web server, and sent a link of the baby’s photo to family and friends. So, I went online and incorporated ezSharing.com Corporation and got the domain names ezSharing.com and FamilyPlex.com so that family members and trusted friends could chat and share photos with one another. Back then, Cineplexes (cinema complexes with multiple screens showing different movies simultaneously) were trendy, with Universal’s Cineplex Odeon merging with Sony’s Loews Theaters to form Loews Cineplex earlier in 1998. I envisioned creating a complex for families to watch photos and videos in their private spaces on FamilyPlex. Working with my wife, a bunch of friends, and neighbors after hours in our garage, we designed and developed the solution, and I started pitching it to Venture Capital (VC) firms in October of 1999. VCs were not interested, especially given Internet privacy and security concerns, and Net Nanny's parent-controlled Internet filtering tools were the hottest topics in the marketplace at the time. One exception to the rule was Tom Toy, a venture partner who sat on the board of directors of Women.com. He liked what he saw, and we ended up selling our technology to Women.com for $18 million in March of 2000. This was four years before Zuckerberg, Saverin, Moskovitz, Hughes, and McCollum launched TheFacebook from their Harvard dorm in February 2004. Proof that Timing, along with some luck, is everything!
In January 2001, I became senior vice president of products and technology at Agile Software. I lead a team designing, developing, and supporting several product lines across multiple technologies following multiple acquisitions, and I oversaw all phases of the SDLC at six development centers across three continents. From 2003 to 2006, I was senior vice president and general manager of the Small and Medium Enterprise Division. During these six years, my work focused on the SDLC as we moved to the Agile Development Methodology and the GAMP V-Model.
In the 10 years that followed, I started my own company, sold it, and then worked at the American University of Beirut (AUB) as Associate Chief Information Officer, with three interim positions: Chief IT Quality and Compliance Officer, Director of IT Customer Success, and Director of Applications Management and Administration. At AUB, I worked on implementing and launching Oracle eBusiness Suite R12 at the university and its medical center. In 2017, I co-founded Beebotte Corporation, a developer of Internet of Things (IoT) platforms, as CEO. Between 2018 and 2021, I was interim CEO of MoneyCam Corporation, where I led the operations, design, and development of an enterprise solution using Amazon AWS, Artificial Intelligence, Blockchain, and Internet of Things (IoT) technologies, accessible from Web, Android, and iOS apps. In 2018, I founded Smart AI Innovations, Inc., focused on developing education management information systems (Epic EMIS) and law enforcement solutions (Epic Law Enforcement) using the latest blockchain, IoT, and AI technologies. The Epic EMIS solution is running at more than 25,000 schools in the Middle East. Since 2012, I have worked as a senior technology and business transformation expert with the World Bank Group and several United Nations agencies, namely United Nations Educational, Scientific, and Cultural Organization (UNESCO), United Nations Children’s Fund (UNICEF), International Labour Organization (ILO), and United Nations Development Programme (UNDP) on digital transformation initiatives and projects in Africa, Asia, and the Middle East.
This book is the culmination of more than 45 years of experience in enterprise solution design, development, delivery, and support. While this book uses the industry-standard SDLC framework as its backbone, the impact of agentic AI extends well beyond the code. Throughout this book, I address the full solution delivery ecosystem — including professional services, deployment, training, and ongoing support — that surrounds and depends on each phase of the SDLC. I have worn every hat and played every role in the SDLC game, and am astonished by how it has evolved over the years. Keep in mind that the core components remain the same, but the way they are designed, planned, and executed is very different, at speeds never seen before. This is a research-grounded, practitioner-tested book on agentic AI in the SDLC, written by someone with diverse experience and real academic rigor. In this book, I will share some milestones and achievements from my career, along with lessons learned. So, I hope you enjoy reading it as much as I enjoyed writing it.