AI-Assisted Development in 2026

AI-Assisted Development in 2026

Claude Code nearly wiped out a production database while racking up an unexpected AWS bill overnight, and that’s just one of many stories from guest Daniel Cannon. As CIO at Delta Systems and CEO of StriveDB, Daniel joins hosts Aaron Marchbanks and Justin Edwards to share what AI-assisted development actually looks like in 2026: real workflows, real mistakes, and why human review is still non-negotiable. What You’ll Learn: * How to structure an AI-assisted development workflow that delivers 10x speed gains without sacrificing code quality * Why Big Design Upfront (BDUF) is making a comeback with AI * The critical difference between letting AI write code vs. reviewing and refining it * How to build safe, controlled AI agents for sensitive workflows * Why context management directly impacts both cost and code quality Daniel Cannon is the Chief Innovation Officer at Delta Systems and Founder and CEO of StriveDB, bringing a wealth of experience in modern web development frameworks and architectures. His expertise spans full-stack development, with particular depth in Ruby on Rails and modern JavaScript frameworks. Daniel's hands-on experience with both traditional and cutting-edge technologies, combined with his ability to evaluate technical trade-offs in practical business contexts, provides valuable insights for organizations navigating technology decisions. This podcast is brought to you by Delta Systems, your one-stop shop for front-end, back-end, and full-stack software development. At Delta, Justin and Aaron share the same philosophy when it comes to clients: they treat people like colleagues, not just customers. Maybe that’s why Delta typically spends years working with the same companies: how many software engineering firms can you say that about? So, if you’ve got a big SaaS project in mind but have no idea where to start, come and get a free scope and estimate from Delta Systems at: https://deltasystems.com/ [https://deltasystems.com/]  Got a burning idea for an episode, or a SaaS question you absolutely must know the answer to? Leave us a voice memo: https://www.speakpipe.com/SaasThatApp [https://www.speakpipe.com/SaasThatApp]  Highlights: * [02:07] Daniel's Daily AI Workflow * [03:18] Big Design Upfront Makes a Comeback * [07:37] The Junior Developer Problem * [10:42] Build the Feature, Learn the Lessons, Nuke It, Start Over * [12:15] Role-Playing Your Way to Better Code Review * [16:25] The Developer Who Let AI Patch Production Unsupervised * [20:28] How a Slack Bot Burned Through $500 on AWS in One Day * [24:07] Meet Athena, Poseidon, and Todd Bots * [31:12] When Claude Goes Wrong * [37:31] AI Tools Are Slot Machines for Developers * [40:11] The Bigger Question Nobody in Tech Wants to Answer We’d love your feedback. Please take a moment to fill out our audience questionnaire: https://forms.gle/DN8hWFDcE9jwvNKo6 [https://forms.gle/DN8hWFDcE9jwvNKo6] Your input helps us shape future episodes and continue bringing you practical, real-world insights into building B2B web applications.

Healthcare AI Is Failing And the Fix Isn't More Data

AI in drug discovery, data-centric machine learning, precision medicine, and building SaaS for life sciences; this episode of SaaS That App covers it all with a founder who’s lived it for a decade. Aaron Marchbanks and Justin Edwards sit down with Dr. Abhishek Jha, Co-Founder and CEO of Elucidata, who makes a compelling case for why the real bottleneck in healthcare AI is data quality, out-of-distribution observations, and the reality of working with small sample sizes where the stakes couldn’t be higher. What You’ll Learn: * How to reframe AI strategy from model-centric to data-centric * Why out-of-distribution detection is existential in healthcare * The managed services pricing model trumps traditional SaaS for AI solutions * How to navigate data sensitivity and IP protection without sacrificing innovation * Why distribution and people problems outweigh technology problems in deep-tech entrepreneurship Dr. Abhishek Jha is the Co-Founder and CEO of Elucidata, a biotech company focused on AI-driven data solutions for life sciences R&D and drug discovery. He has over 20 years of experience at the intersection of life sciences, data science, and machine learning. Abhishek co-founded Elucidata in 2015, launching Polly as a cloud-based SaaS platform that harmonizes multi-omics, clinical, and real-world data into ML-ready formats, speeding data preparation by 10x with high accuracy via LLM-powered curation. This podcast is brought to you by Delta Systems, your one-stop shop for front-end, back-end, and full-stack software development. At Delta, Justin and Aaron share the same philosophy when it comes to clients: they treat people like colleagues, not just customers. Maybe that’s why Delta typically spends years working with the same companies: how many software engineering firms can you say that about? So, if you’ve got a big SaaS project in mind but have no idea where to start, come and get a free scope and estimate from Delta Systems at: https://deltasystems.com/ [https://deltasystems.com/]  Got a burning idea for an episode, or a SaaS question you absolutely must know the answer to? Leave us a voice memo at: https://www.speakpipe.com/SaasThatApp [https://www.speakpipe.com/SaasThatApp]  Highlights and YouTube Chapters: * [02:21] Everyone's Obsessed With Models. He Says They're Looking at the Wrong Thing. * [03:17] He Spent 30 Years Training for a Career He Never Started * [06:34] Profitable on Day One, No VC Required * [09:17] Facebook Has 3.5 Billion Users. The Best Clinical Trial? Tens of Thousands. * [15:09] They Tried SaaS Pricing. It Broke. * [16:34] Why a Single Security Glitch Is Existential * [22:04] The Most Brutally Honest Take on Healthcare AI * [25:02] The #1 Mistake Scientists-Turned-Founders Keep Making * [27:16] “No Excel Model Justifies This” Episode Resources: * Dr. Abhishek Jha on LinkedIn [https://www.linkedin.com/in/abhishek-jha-leading-data-centric-solutions/] * Elucidata Website [https://www.elucidata.io/] * Aaron Marchbanks on LinkedIn [https://www.linkedin.com/in/aaronmarchbanks/] * Justin Edwards on LinkedIn [https://www.linkedin.com/in/iterative-justin-edwards/] * Delta Systems Website [http://www.deltasystems.com/?utm_source=YouTube&utm_medium=Video&utm_campaign=SaaSThatApp]   We’d love your feedback. Please take a moment to fill out our audience questionnaire: https://forms.gle/DN8hWFDcE9jwvNKo6 [https://forms.gle/DN8hWFDcE9jwvNKo6] Your input helps us shape future episodes and continue bringing you practical, real-world insights into building B2B web applications.

Your Product Isn’t Ready for AI Until You Fix These Software Mistakes

From governed agents to token cost management and smarter architecture decisions, fractional CTO shares practical AI strategy for SaaS startups. In this episode of SaaS That App - Building Tech-Enabled Businesses, Garrett Fritz, a Partner and Chief Technology Officer (CTO) at MetaCTO, joins Aaron Marchbanks and Justin Edwards to dive into his two guiding principles of software development: “Nobody Reads” and “The Previous Guy Wasn't an Idiot.” What You’ll Learn: * Why half the startups claiming to be AI companies are really just fancy if-statements * How the coming shift to usage-based token pricing is about to reshape who can afford to vibe-code their way to a product * The real cost of runaway API keys * Why non-technical founders should test the waters with AI Garrett Fritz is a Partner and Chief Technology Officer (CTO) at MetaCTO, a firm focused on building, growing, and monetizing mobile apps. He specializes in helping non-technical founders and business leaders develop scalable mobile and AI solutions through digital transformations. As Partner and CTO at MetaCTO, Garrett combines his MIT aerospace engineering background with extensive entrepreneurial experience to deliver exceptional mobile and AI solutions for clients. This podcast is brought to you by Delta Systems, your one-stop shop for front-end, back-end, and full-stack software development. At Delta, Justin and Aaron share the same philosophy when it comes to clients: they treat people like colleagues, not just customers. Maybe that’s why Delta typically spends years working with the same companies: how many software engineering firms can you say that about? So, if you’ve got a big SaaS project in mind but have no idea where to start, come and get a free scope and estimate from Delta Systems at: https://deltasystems.com/ [https://deltasystems.com/]  Got a burning idea for an episode, or a SaaS question you absolutely must know the answer to? Leave us a voice memo at: https://www.speakpipe.com/SaasThatApp [https://www.speakpipe.com/SaasThatApp]  Highlights and YouTube Chapters: * [01:56] He Left MIT Rocket Science for Android Apps? * [06:30] Principle #2: “The Previous Guy Wasn’t an Idiot” * [09:48] Principle #1: “Nobody Reads” * [12:17] Write More Docs, But Not for Humans * [15:34] “AI Booked Us a $400 Dinner We Couldn't Eat” * [17:13] Your Vibe-Coded Prototype Might Cost You $50K in Tokens Alone * [22:34] The Only AI Strategy That’s Actually Making Enterprises Money * [27:18] 10 Devs, 20 Different Outputs * [31:17] Try AI, But Don’t Back Out of the Driveway Episode Resources: * Garrett Fritz on LinkedIn [https://www.linkedin.com/in/grfritz/] * MetaCTO Website [https://www.metacto.com/] * Aaron Marchbanks on LinkedIn [https://www.linkedin.com/in/aaronmarchbanks/] * Justin Edwards on LinkedIn [https://www.linkedin.com/in/iterative-justin-edwards/] * Delta Systems Website [http://www.deltasystems.com/?utm_source=YouTube&utm_medium=Video&utm_campaign=SaaSThatApp] We’d love your feedback. Please take a moment to fill out our audience questionnaire: https://forms.gle/DN8hWFDcE9jwvNKo6 [https://forms.gle/DN8hWFDcE9jwvNKo6] Your input helps us shape future episodes and continue bringing you practical, real-world insights into building B2B web applications.

Why Most SaaS Companies Fail to Scale in the AI Era

Scaling SaaS isn’t about hiring more salespeople or chasing vanity growth, it’s about brutal honesty, focus, and timing. In this episode of SaaS That App, four-time founder Geoff McQueen joins Aaron Marchbanks and Justin Edwards to to unpack the realities of SaaS scaling, from navigating the founder journey and making the services-to-product transition, to understanding venture capital dynamics, company exits, and what it truly takes to achieve sustainable growth. From pricing trade-offs to exit timing, Geoff shares hard-earned lessons on building, scaling, and selling SaaS in a market where barriers are low but the stakes have never been higher. What You’ll Learn: * Why trying to run services and product businesses simultaneously kills focus and what to do instead * Why hiring more sales reps doesn’t fix broken growth systems * How pricing shifts when moving from high-touch services to scalable SaaS * The difference between confident leadership and false certainty and why it matters * How to recognize and capitalize on your company’s peak exit window * Why AI is lowering barriers to entry while dramatically increasing competition * What the next wave of SaaS disruption means for legacy, overpriced platforms  Geoff McQueen [https://www.linkedin.com/in/geoffmcqueen/] is a four-time founder, currently CEO of Ascendius Inc. [https://www.ascendius.co/] and creator of the product, WorkSights AI, a platform focused on continuous performance intelligence for modern leaders. With over 20 years of experience building and scaling technology companies, Geoff has led multiple ventures from inception to exit. He previously co-founded and served as CEO of Accelo, a leading client work management platform that scaled to over $14M in ARR before its exit to private equity in 2024. Earlier in his career, he built a successful digital agency, Internetrix, growing it from a solo operation into a large-scale business serving clients ranging from small companies to government organizations. Geoff is also an active investor and mentor, particularly through StartMate, one of Australia’s leading startup accelerator communities. Now based between Denver and San Francisco, he continues to build and advise companies at the intersection of SaaS and AI. This podcast is brought to you by Delta Systems, your one-stop shop for front-end, back-end, and full-stack software development. At Delta, Justin and Aaron share the same philosophy when it comes to clients: they treat people like colleagues, not just customers. Maybe that’s why Delta typically spends years working with the same companies: how many software engineering firms can you say that about? So, if you’ve got a big SaaS project in mind but have no idea where to start, come and get a free scope and estimate from Delta Systems at: https://deltasystems.com/ [https://deltasystems.com/] Got a burning idea for an episode, or a SaaS question you absolutely must know the answer to? Leave us a voice memo at: https://www.speakpipe.com/SaasThatApp [https://www.speakpipe.com/SaasThatApp] Highlights: * [00:00:00] The Founder Identity Trap That Destroys Companies * [00:01:54] From 3AM Milk Runs To SaaS Founder * [00:04:36] Moving From Services To Product: The Neverending Dilemma * [00:08:41] Why Bootstrapping Feels Brutal (Even with 6-Figure LTV) * [00:11:13] How Frustration Can Fuel A Startup * [00:17:37] Leading Through Uncertainty Without Losing Hope * [00:20:04] Hiring More Salespeople Will NOT Boost Your Revenue * [00:24:13] The Moment You Raise VC, You’ve Already Sold Your Company * [00:33:04] Building WorkSights AI And The Future Of SaaS    Episode Resources: * Geoff McQueen on LinkedIn [https://www.linkedin.com/in/geoffmcqueen/] * Ascendius Inc. Website [https://www.ascendius.co/] * Aaron Marchbanks on LinkedIn [https://www.linkedin.com/in/aaronmarchbanks/] * Justin Edwards on LinkedIn [https://www.linkedin.com/in/iterative-justin-edwards/] * Delta Systems Website [http://www.deltasystems.com/?utm_source=YouTube&utm_medium=Video&utm_campaign=SaaSThatApp]  We’d love your feedback. Please take a moment to fill out our audience questionnaire: https://forms.gle/DN8hWFDcE9jwvNKo6 [https://forms.gle/DN8hWFDcE9jwvNKo6] Your input helps us shape future episodes and continue bringing you practical, real-world insights into building B2B web applications.

Hype or Game Changer? How Developers Are Actually Using Claude Code

What does day-to-day software development actually look like when AI coding tools like Claude Code, Codex, and LLM-powered agents run the show? In this episode of SaaS That App, Aaron Marchbanks and Justin Edwards are joined by Nolan Alimonti, Senior Architect and Team Lead at Delta Systems, to share how Claude Code and Codex are reshaping their entire workflow, why code quality still matters when AI generates 10x the output, and the critical skills you need to manage AI agents effectively. What You’ll Learn: * How to treat AI coding tools as coworkers * Why code review and validation are now your critical bottleneck * How to use minimal, focused CloudMD instructions instead of massive documentation * Why pairing multiple models catches bugs that single models miss * How to prevent AI from pushing bad patterns into your codebase Nolan Alimonti is a Senior Architect and Team Lead at Delta Systems, where he tackles complex client projects by uniting front-end speed with backend clarity. With deep experience in React and Ruby on Rails, he’s known for simplifying messy codebases, improving scalability, and steering apps toward cleaner, maintainable architecture. A hands-on specialist in modern Rails development, Nolan has led significant refactors, helping teams migrate from Turbolinks-era code to sleek, serviceable Turbo and Stimulus builds. His goal is  to make apps snappier and smoother, while keeping the stack lean and intentional. This podcast is brought to you by Delta Systems, your one-stop shop for front-end, back-end, and full-stack software development. At Delta, Justin and Aaron share the same philosophy when it comes to clients: they treat people like colleagues, not just customers. Maybe that’s why Delta typically spends years working with the same companies: how many software engineering firms can you say that about? So, if you’ve got a big SaaS project in mind but have no idea where to start, come and get a free scope and estimate from Delta Systems on their website [http://www.deltasystems.com/?utm_source=YouTube&utm_medium=Video&utm_campaign=SaaSThatApp]. Got a burning idea for an episode, or a SaaS question you absolutely must know the answer to? Leave us a voice memo: https://www.speakpipe.com/SaasThatApp [https://www.speakpipe.com/SaasThatApp]. Highlights: * [00:00] From Writing Code to Managing the Machine That Writes It * [01:59] Ditching the Browser for Claude Code and Codex * [04:25] AI as a Coworker You Delegate To * [06:01] Why the First Run Is Never Right * [09:11] The Front-End Design Skill That Changed Everything * [12:18] More Output, Same Workload * [16:16] Pitting Models Against Each Other for Better Code * [18:13] The New Bottleneck: Code Review, Not Code Writing * [21:08] Why Senior Devs Get 10x Value and Juniors Get 3x * [27:36] The Laziest Thing Nolan Got Away With Episode Resources: * Aaron Marchbanks on LinkedIn [https://www.linkedin.com/in/aaronmarchbanks/] * Justin Edwards on LinkedIn [https://www.linkedin.com/in/iterative-justin-edwards/] * Delta Systems Website [http://www.deltasystems.com/?utm_source=YouTube&utm_medium=Video&utm_campaign=SaaSThatApp] We’d love your feedback. Please take a moment to fill out our audience questionnaire: https://forms.gle/DN8hWFDcE9jwvNKo6 [https://forms.gle/DN8hWFDcE9jwvNKo6] Your input helps us shape future episodes and continue bringing you practical, real-world insights into building B2B web applications.