Jim Riley worked as a pilot for California Air National Guard. He flies a fighter jet for fun. He also started several software companies, including Learn.com, which he sold in the late 1990s.

But by far the hardest thing he’s ever done in his life?

Starting a health-technology company.

“What was supposed to be a fun one or two-year project turned into a personal odyssey with a good part of my net worth invested in it,” said Riley by phone.

Riley is far from alone. His experience is so common among the growing hoards of tech workers moving into health care that some hospital executives even have a name for it: The “pit of despair.”

Aaron Martin, chief digital officer at Providence Health, said he sees this happen frequently. Martin used to work at Amazon, and he’ll often recruit engineers from his former employer and other tech companies to his team. He likes the fresh perspective, and there’s plenty of work to do to digitize the medical system.

According to Martin, many of these developers will first see a massive opportunity to bring health care into the 21st-century. That excitement turns to bitterness after they begin to understand the complexities of health care with its entrenched interests, regulatory hurdles, long sales cycles and more.

“There’s this pit of despair when people realize that this stuff is really hard and complicated,” said Martin. “And the penalty for screwing up is high.”

Riley almost gave up on his company many times.

Like many in the technology sector, he decided to transition into health care after a personal brush with the system. His father, who had just had a heart attack, needed to be transferred to another hospital. The nurses told Riley that they’d need to procure his father’s medical records first.

Riley imagined a process where those were batched and transferred electronically in a matter of minutes. Instead, workers photocopied paper recrods, crammed them into a giant manila file and literally strapped it to his father’s gurney.

“This was 2009, a time of Facebook and the iPhone,” he said.

Riley, who noted that many hospitals had not yet moved to the cloud, figured that he could use his software skills to build a cloud-based alternative to the on-premise medical record systems. He figured it would be an easy enough service to build with a few developers and sell to hospitals.

This period of optimism did not last long.

Riley quickly came up against intense competition and cultural resistance. Few vendors would partner with him. Hospitals had already invested millions or even billions of dollars into on-premises medical record systems, which they were fully locked into. So he ended up being out more and more functionality, and pouring millions of his personal funds into the service.

Years later, he’s finally seeing a light at the end of the tunnel.

“Every time you’d overcome a roadblock, there would be two more down the road,” he said. “It was almost overwhelming.”