Mark Collins, co-founder and Executive VP of Technology at Intechrum, joins AI InterConnect with a refreshingly practical view of AI. Intechrum helps organizations modernize business processes through software development, workflow automation and, increasingly, AI-enabled solutions. Mark's rule is simple: understand the problem first, then decide whether AI makes sense.
From enterprise software to his own company
Mark spent years in enterprise software development, most recently on a large HR system for the state of California. When the company behind it decided to leave that business, the team spun off to keep serving the state, and Mark started Intechrum with his business partner.
In consulting he often saw organizations buy a tool first and then look for a problem to solve with it. Intechrum works the other way round: custom software built to solve a specific problem the client already has.
Understand the business first
Non-technical leaders often arrive with a solution in mind without knowing what is possible. Mark's first questions are what keeps them up at night, why they are in business and how they add value to their community.
Only then does the team choose technologies and design a solution, making sure it delivers a return on investment rather than being a shiny new thing.
Augment people, do not replace them
Employees have a valid fear: some companies have implemented AI and laid people off. Mark's approach is to use AI to take over the work people hate, such as emails, forms and reports, so they can do work that matters and become more engaged in the business.
People still need to oversee the AI, which makes mistakes. Even in software development, where AI can now write much of the code, someone has to understand the process well enough to direct it, spot-check it and keep it on track.
Small automations with fast payback
Custom software once cost $100,000, $200,000 or $300,000, out of reach for many small companies. With AI, Mark can now build for a single person's task. One example: automating customer intake, with AI extracting details from a PDF and entering them into the system, built within days or a couple of weeks.
If a project of around five thousand dollars gives a person back six hours a week, the ROI is obvious. He also sees very small teams of two or three people competing with much larger companies, because AI lets them scale without matching headcount.
Government, change and a founder's gaps
In government, the hardest part of modernizing is rarely the technology. Regulation slows things down, many people want a say, and people get comfortable: the same staff who resisted a system later resist replacing it.
As a technical founder, Mark finds marketing and sales the hardest part, alongside long government sales cycles. He relies on a business partner for project management and even sets up AI personalities as a kind of advisory board, while watching for hallucinations and rabbit holes. Intechrum is now shifting toward private-sector businesses, with targeted workflow projects of around 5k or 10k.
What to do on Monday morning
Mark's advice to a business leader who knows AI matters but not where to start: run an assessment. Understand where the business adds value, which pain points are repetitive enough to automate, and what systems already exist.
Then prioritize by ROI, set guardrails rather than telling everyone to start using AI, implement one project at a time and show results. Success, for him, is hearing that a task that took two months now takes ten minutes.
Watch the full conversation
Watch Mark Collins explain how to start AI projects from real business problems and prove ROI.
AI InterConnect hosts conversations with founders, investors and operators building with AI. Explore more episodes on the AI InterConnect podcast page.