Home Our Work Expertise How We Work About Resources Join Our Team

Insights

Opinionated writing on AI, machine learning, and the business of technology, from our team. No hype, no paid placement. Just the view from 25+ years of doing this work.

June 18, 2026 · Rebecca Joy· LinkedIn

Your Competitor Didn't Beat You With Better AI

They beat you with better clarity. AI amplifies whatever direction you're already moving, so if your strategy is fuzzy, it just moves you faster in the wrong direction. Rebecca compares today's AI rush to the late-'90s website scramble: the companies that rushed in without a plan paid twice, once to build it wrong and again to rebuild it. She's watched the same pattern repeat with AI: a green-lit tech team, six months of work, and a solution looking for a problem because the data wasn't ready and the use case was never defined. The hardest part is rarely the technology. It's getting the business side and the tech side in the same room, speaking the same language, long enough to build the same thing before a line of code gets written. That gap, alignment before automation, is the work behind both Saigon A.I. and QCoda.

Read on LinkedIn

June 11, 2026 · Andrew Schwabe· Hackernoon

What the Heck Is a Model Ladder? And Why Do I Need to Use One?

When you build agents on local, open-source LLMs, a model can simply disappear out from under you, and your agent grinds to a halt. Andrew's answer is the model ladder: an ordered-fallback pattern that tries your preferred model first, then climbs down to alternatives so the system keeps running instead of failing. He walks through the pattern in pseudocode and Python, and lays out the four decisions that make a ladder actually work, from how you rank models to when you give up and escalate. A practical playbook for anyone running AI in production who can't afford a single point of failure.

Read on Hackernoon

June 4, 2026 · Rebecca Joy· LinkedIn

What AI Actually Costs

Everyone asks Rebecca about token costs. It's the wrong question. After one client's capped $200/month budget ballooned to $1,000 by month four, she lays out the real price of AI: the costs no vendor puts in a proposal. Data privacy that's a governance problem, not a technical one. The hunt for clean "Golden Data" she's never once seen in the wild. Change management that's underestimated every time. The people hours quietly absorbed into salaries. And the failed pilots nobody tracked well enough to even tally. The CFO ends up holding all the bills, and the question nobody wants to answer: was it worth it? The tool is not the investment, Rebecca argues. The transformation is, and it has a price that goes well beyond the token bill.

Read on LinkedIn

May 21, 2026 · Andrew Schwabe· Hackernoon

Comprehension Debt Found Me

Andrew noticed he was typing "please" to his AI, asking about commits on a feature he had led himself. That reflex turned out to be a signal. When AI writes the code, the tests pass and the PR looks fine, but the developer never fully internalizes the why. Andrew calls this comprehension debt: not a failure of the AI, but a quiet erosion of your own understanding. It's the layer of human failure that nobody is studying because it doesn't show up in any tool's error log. He built QCoda to address it: a place outside your head where the reasoning behind decisions gets written down as the work happens. The bigger warning is about autonomous agents: more autonomy for the AI means less agency for you. The answer isn't slowing the AI down. It's staying in the room.

Read on Hackernoon

May 21, 2026 · Rebecca Joy· LinkedIn

The Efficiency Trap

AI is supposed to save us time. So why are we more exhausted than ever? Rebecca draws on 30 years in tech, and the economics of the Jevons Paradox, to argue that we're not using AI to work less. We're using it to take on more than we ever could before. Only 21% of employees use the time AI saves for anything other than more work. The rest put it straight back into the job. And there's a second, quieter cost: the thinking we're outsourcing is the thinking that keeps us sharp. The professionals who thrive won't be the ones who use AI the most, they'll be the ones who know when not to.

Read on LinkedIn

May 12, 2026 · Andrew Schwabe· Hackernoon

AI Scrapers Stole Our Job Posting and Started Collecting Applications We'll Never See

When Saigon A.I. posted a job opening on their own website, it appeared on LinkedIn within hours, posted by no one on their team, accepting applications that would never reach a real inbox. As Andrew Schwabe explains in this piece, this is happening across the internet right now: AI scrapers and job aggregators harvest career pages, repost listings on trusted platforms, and collect candidate resumes that disappear into a black hole. Applicants spend time tailoring cover letters for jobs that were never going to be reviewed. Andrew walks through exactly how it works and how companies can fight back using free Cloudflare tools to lock down their hiring pages in under an hour. It's also a big part of why we're building Talential: because real talent deserves a direct line to real opportunities.

Read on Hackernoon

May 7, 2026 · Rebecca Joy· LinkedIn

Beyond Vibe Coding: What It Actually Takes to Build AI Systems Businesses Can Trust

Three months of vibe coding a QuickBooks replacement revealed the gap AI tools create: fast starts, slow finishes, and code no junior developer can maintain. Rebecca argues the real problem isn't prompting. It's alignment between business intent and technical execution, and why that gap can't be patched with a better model.

Read on LinkedIn

May 5, 2026 · Andrew Schwabe· Hackernoon

Streaming Faster Made Our LLM Hub Slower

Wiring up per-token streaming to an LLM inference mesh felt great with one user, and fell apart with three. A deep dive into the adaptive batching logic, control-theory feedback loops, and the one measurement mistake that would have caused a death spiral.

Read on Hackernoon

April 29, 2026 · Andrew Schwabe· Hackernoon

I Trusted AI With My Code… And It Cost Me Half a Day and a Fistful of Tokens

A first-person account of what happens when you defer too much to an AI coding assistant. Claude confidently argued against a proven approach, Andrew followed it, and a half-day of failed tests later the original design was reinstated.

Read on Hackernoon

April 16, 2026 · Andrew Schwabe· Hackernoon

We Built a Local Model Arena in 30 Minutes — Infrastructure Mattered More Than the App

Andrew and the team built a local LLM arena in half an hour using LLMesh, and the lesson wasn't about the app. The infrastructure choices underneath determined everything: speed, reliability, and whether the model comparisons meant anything at all.

Read on Hackernoon

April 14, 2026 · Andrew Schwabe· Hackernoon

Nobody Is QA Testing Their LLM Apps (That's Going to Be a Problem)

Most teams ship LLM and RAG applications with no real test suite, and are discovering the consequences in production. Andrew presents a six-layer testing framework that addresses the full stack of quality concerns.

Read on Hackernoon

March 19, 2026 · Andrew Schwabe· Hackernoon

Alignment-Driven Development: The Orchestration Framework AI-Paired Coding Forgot to Build

AI coding assistants are fast, but fast at what? Andrew argues that the missing layer in every AI-accelerated team is alignment: a documented, auditable process that ensures every decision stays connected to actual goals.

Read on Hackernoon

February 23, 2026 · Andrew Schwabe· Hackernoon

Urgency as Office Culture (duh, AI makes it worse)

AI isn't creating urgency culture. It's amplifying the one that was already there. Andrew examines how leaders can resist the always-on hype cycle and address the anxiety that AI acceleration is making worse.

Read on Hackernoon

November 17, 2025 · Andrew Schwabe· Hackernoon

Calling All Brains! GenAI Misconceptions

The most dangerous GenAI myths aren't about what the technology can't do. They're about what people believe it already does. Andrew breaks down the most common LLM misconceptions and what responsible AI adoption looks like.

Read on Hackernoon

October 10, 2025 · Andrew Schwabe· Hackernoon

Doubling Down on Classic AI: Why I'm Betting on Science Over Spectacle in Business

While generative AI dominates the headlines, the case for classic AI in business has never been stronger. Andrew argues that reproducibility, scientific rigor, and reliability are exactly what enterprise AI needs to earn trust.

Read on Hackernoon

October 2024 · Andrew Schwabe

Living With or Without Big "A" Algorithms

A conscious effort to reclaim control from algorithms and targeted advertising, and what a family project hacking old iPods taught Andrew about intentional technology use.

September 2022 · Tram

Our Company Retreat in Three Words

Three days in Vung Tau. The whole team, away from screens and schedules, captured in three words: Different, United, Enjoyable.

March 2022 · Andrew Schwabe

What Does a Career Look Like for a Serial Entrepreneur?

The serial entrepreneur model: juggling multiple companies at different stages, managing risk, and always showing up as a great employee first.

January 2022

HCMC Supports AI Research — And So Do We

Ho Chi Minh City launches an initiative to support AI research and development. Saigon A.I. opens internship opportunities for students entering the field.