Created: 08/17/2026 12:00 PT
Last updated: 08/20/2026 17:13 PT
November 2023 through April 2025
OptHealth's technology stack was:
Following a period of resource constraints, OptHealth again had the resources to reinvest in their technology stack, but they had difficulty finding someone willing to lead the engineering team out of the hole they found themselves in.
OptHealth had a completely up-to-date technology stack, with a rebuilt core, a highly functional engineering team that spent less than 10% of their time on bug fixes, and the original engineer was able to continue working and has called me "the best mentor they ever had" and "changed the trajectory of their life". We also maintained a 99.99% uptime (measured via automated health checks against critical paths post-implementation; no baseline tooling existed prior) for the core web application experience, after the first six months of improvements were implemented.
For OptHealth, the engineering organization and technical stack were no longer a blocker to growth or fundraising with the CEO describing engineering as the best-run org in the company.
While I was overall very happy with the end result for OptHealth and the team, there were a couple of major areas for improvement. I was less satisfied with the toll it took on me and there were some specific process transfer failures.
Reflecting on the experience, there were clear areas where I took on extra work and responsibility, that were not necessary for achieving the outcome, and made the work less sustainable for me and less useful to making the team and company better.
At times, we were under-resourced, especially in cases where the scope was changed partway through a project or when it was not well defined enough at the start. While I did sometimes flag these issues, instead of clearly highlighting to the executive team the additional work it created, I would often just help with carrying out the work, to make up the difference. I know I did this because dealing with the root causes for why those issues were happening seemed like it would take more resources than I had.
While the team I built met my goals, I did fail to successfully transfer the process I used to the rest of the organization to utilize for future hires. The failure was ultimately on me for not securing more clarity and alignment from the other people involved, which led to me not clearly instituting an easily reproducible process, in this case.
I believe the first failure was because I did not set clear boundaries and escalate when I should have. The second failure was because I did not make enough of an effort to ensure there was clarity and alignment in my understanding. I believe I would have had the resources to properly deal with the issues had I been more proactive initially. In future engagements, I intend to go into them with clearer boundaries set and communicated ahead of time and always check with myself to ensure I really understand things at the level I need to. I believe that would give me the space and energy to do a better job in the future.
I was hired as the Head of Engineering and reported directly to the CEO, with the engineering team reporting to me. Our team also collaborated with a Project Manager who also reported to the CEO.
I led the entire engineering effort. This included: evidence gathering, analysis, decision support, recruitment and hiring, high-level project planning, feedback on major architectural choices, performance reviews, critical code reviews, establishment of engineering processes such as CI/CD and code reviews, and overseeing the daily activities of team execution.
OptHealth is a seed-funded telehealth tech startup based in San Francisco, California and founded in 2020. Due to resource constraints, the original development team was not able to stay with the company. There departure was followed by a long period in which the company were unable to invest in the technology they had built, which led to high levels of technical debt.
When joining, the technical stack consisted of:
core web-app:
integrated with
an iOS app built by an outside agency utilizing React Native
software that had to be HIPAA-compliant
None of the software packages in any part of the stack or iOS app had been updated since 2020.
To begin, I surveyed:
The scope and health of the current technology stack.
The things the engineer was spending their time on.
The processes utilized by the engineering, operations, and planning teams.
The psychological state of people involved.
The expectations of the executive team.
The historical record in relation to: bugs, technical debt, and engineering resource utilization.
The most immediate problems I discovered were:
mis-alignment between executive team expectations and on-the-ground reality leading to constant distress on all sides
engineering resources being spent on short-term fixes while increasing long-term technical debt
engineering resources being spent on handling operational issues
This set of problems clearly indicated a "Fixes That Fail" negative feedback loop where the system continued to degrade due to a prioritization of short-term fixes over long-term stability.
I identified three significant, immediate leverage points:
The misalignment of expectations needed to be resolved.
Engineering resources needed to be reallocated from operational issues to engineering and development.
Other company resources were available to handle operational issues, so a training and hand-off process was needed to move top-level operational support from engineers to other teams.
The executive team needed to be made fully aware of the current state of the technical debt and what actions they could take to move towards a positive feedback loop.
At the start I:
Interviewed the engineer, project manager (who had recently joined full-time), and the executive and product teams.
Performed a technology and code analysis and produced a decision support report which included:
Executive score summary for each category, for example (this is a methodology example, actual assessment is confidential)
| Category | Score (Out of 10) | Goal |
|---|---|---|
| Security | 2 | 8 |
| Reliability & Quality of Service | 3 | 4 |
| Integrations | 1.5 | 7 |
| Eng. Team Performance | 4 | 10 |
| Infrastructure | 2 | 10 |
| Compliance | 5 | 10 |
"score" is the current situation calculated based on the report details and the "goal" is the level I determined OptHealth should be at for the stage of the company and industry. The "goal" values would increase as the company scaled and/or raised money.
Estimated investment needed to reach each "goal" as expressed in "engineering person hours".
A high level overview of what I considered to be immediate risks and long-term risks.
And the rest of the report was a thorough breakdown of the specific reasons for each score and where they come from.
I then worked with the CEO and other stakeholders to determine priorities and desired timelines. After that, we worked together to identify a resource plan based on priorities and runway. Following a shared understanding on direction and resources, I developed an execution plan based on the expected change in resources and secured approval from the executive team.
At the same time, I worked with the engineer to codify how they handled operational issues, and we handed off that responsibility to other teams. I also prioritized an initial fix for the development environment and first pass at creating checklist processes for critical tasks like deployments.
And, most importantly, I established a psychologically safe environment so people could work in a healthy manner and provide necessary feedback without fear. This would be critical to being able to get good feedback in the future and create a more robust and adaptable engineering system.
After ensuring the immediate negative feedback loop was broken, I worked towards establishing a positive feedback loop where the company's capabilities would be expanded over the long term and the engineering team and software stack would be well positioned for business growth.
I approached building the engineering team and processes with the goal of creating a system in which the engineers could sustainably produce high-quality output while operating within a fast-paced and ever-changing startup environment while also keeping the overall cost as low as possible.
I instituted a number of policies and processes to reduce the amount of engineering resources that would be spent on low-leverage tasks and increased the amount of time available to spend on high-leverage tasks. This included:
Leading a more robust and sustainable setup to unify the development, staging, and production environments.
The goal was that production could become a stable environment and testing and validation could take place in the environment with the shortest and cheapest feedback-loop (generally development and sometimes staging, depending on the integration environment required). This was accomplished with a series of smaller tasks, carried out by the engineering team, to resolve data and integration differences and automate the package installation and data management within the development environment.
Building on the initial deployment process improvements to create a lasting process that would ensure human mistakes were eliminated, deployments were validated, and failing deployments could be quickly identified and easily rolled back.
Creating a fully developed code review process that would ensure a trajectory towards higher-quality solutions without adding overhead that the company could not afford.
This was accomplished by classifying changes by risk level with risk level being determined by: operational risk, security risk, and architectural risk.
Creating a clearly defined task description and assignment process so people could more effectively focus on carrying out tasks.
In addition to the policies and processes, I coordinated with the executive team, project manager, and the rest of the organization to expand the engineering team by two full-time members and a variable number of additional contractors to provide supportive services such as additional engineering effort and UX design.
I led the hiring and recruitment process, provided compensation targets to the CEO, and negotiated with hires.
I focused on building a team that would function highly together and prioritized "team players" over raw technical ability. This created a team that grew in capability and cohesiveness over time and would be resilient and adaptable to changes.
Beyond building the engineering team we also needed to carry out a technical modernization. Throughout this process, my goal was not just to succeed at updating the core technology stack but also to minimize risk and transfer the skills to the rest of the engineering team.
I led a migration from the DrChrono EHR to the Healthie EHR.
This was a massive project as the EHR integration underpinned much of the functioning and structure of the software application but we determined this migration would make it much easier to scale and would reduce long-term engineering and operational costs due to Healthie being a better structural fit for the type of application that OptHealth was.
While I led this process, I focused on developing the capability of the rest of the team to do so on their own in the future. I did this by progressively handing off the tasks I was performing as the other team members seemed ready to handle it. Then I would provide feedback until they were able to do it on their own.
I oversaw the original engineering team member perform an integration with HealthGorilla.
Even though this engineer had started at the company with relatively little experience, by the time we needed to do this integration, I felt they were ready to lead it themselves. I let them take the lead on the entire project. I provided feedback and mentorship. There were some areas for improvement, but during a later integration with a health tracking provider, this engineer applied their learning from the HealthGorilla integration and was able to completely handle the entire project successfully on their own.
Another massive project we needed to accomplish was: updating all software packages to be current and implementing processes to ensure they stayed current, including:
Hundreds of NodeJS packages along with underlying infrastructure, like Postgres.
NodeJS from v12 to v22 (including many breaking code changes)
Before I joined, other engineers whom the company consulted thought this would be impossible with the resources a seed-funded startup like OptHealth had available to it. This process took about four months. I worked with the engineering team to build out an update plan that would allow things to be updated incrementally and validated at every step. It was not a fun task for the senior engineer who carried out the upgrade, but the process worked and the upgrade was accomplished without any critical bugs being introduced.
After the full upgrade was complete, I initiated the effort to ensure the packages would stay up-to-date by working with the engineering team to create an automated GitHub process that would check for package updates, apply them, run the test suites, and create a PR ready to merge, including the test results and a list of major or breaking changes.
I also led the engineers on:
Creating a CI/CD pipeline.
Improving application code and internal data architecture.
And I ensured the engineers learned my process for evaluating and creating well-architected code.
Instituting an automated testing platform and requirements.
Developing a core integration test suite.
Creating a separate operations and administration platform to offload tasks from engineering to the operations team.
While there were multiple large and "high effect" projects necessary to reduce the technical debt, the biggest thing I emphasized and taught the team was how to carry out large changes like this with low risk. I taught them multiple techniques and when and how to apply them: such as developing systems in parallel, breaking systems down into smaller parts that can be worked on independently, and ensuring consistent inputs and outputs between both versions. This ensured that the engineers would not get overwhelmed, that another engineer could take over if they left, and that projects would get stuck at "80%".
In addition I:
Mentored and guided the less experienced engineers to increase their technical and team capabilities.
For the most senior engineer, this meant mentoring them and training them on becoming the team tech lead.
Led and hired a contractor team to build a new, native iOS app.
Provided engineering team status and progress updates to the executive team.
I also trained the most senior engineer on the system I used to do this and gave them experience doing it as well.
Created and presented regular updates on technology stack status and goals.
This was a continuation of the original "decision support" report I created initially. This allowed the executive team to transparently track the engineering team's progress and enabled them to easily make decisions on product direction and resourcing. The report provided a robust method of transferring engineering-related consequences into executive decision-making language and made it easy to point out the effect of proposed changes in terms of their impact on the score.
In conclusion, I was able to help OptHealth break out of a negative feedback loop and establish a positive feedback loop while also ensuring the engineering organization and technical stack were no longer blockers to growth or fundraising. I also created a system in which no one person was the "linchpin" for any task and the team had the processes and training to be adaptable to the swiftly changing needs of a startup while also being able to work sustainably while also leaving individuals more capable than when I began working with them.
By far, the biggest challenge was not always creating the conditions that would allow me to work sustainably and successfully transfer processes to the rest of the organization. I look forward to being able to take on another similar challenge, but with a better understanding of how to do it in a more sustainable way for myself and ensuring I always seek the alignment and clarity I need to transfer processes to the organization successfully.