Felipe Lachowski · Platform Engineering Leader

I build systems that run themselves,
and teams that outgrow me.

Running software was always the expensive half, so that is the half I automate. On my own product I moved repair across the line, and I own the envelope it works inside, the budget it spends, and the switch that stops it.

A platform-engineering leader who builds AI-first systems that run themselves: they watch their own health, repair what their monitoring detects, and escalate the moment they are out of their depth, inside an envelope a person sets. 15+ years making hard things work, from gas-turbine control software to cloud platforms at Citrix and NBCUniversal.

Citrix → NBCUniversal → Vertex

A personal page, not a pitch. I’m a Platform Development Manager at Vertex; the opinions here are mine, and so are the systems I own end to end on my own time.

Felipe Lachowski
Ask-Felipe the AI

Hi, I'm Ask-Felipe, an AI assistant. Ask about his background, how he works, or what he's built. I answer only from his verified profile.

Ask-Felipe is an AI · answers come only from Felipe’s verified profile · messages may be saved so he can follow up. Privacy
The arc

From gas turbines to multi-agent AI.

I started in aerospace and gas-turbine engineering, writing real-time turbine combustion-tuning software. I carried that systems mindset into software, site reliability, and platform leadership, and now into AI-first systems that have to stay reliable in production, which is exactly where most AI stalls.

At a glance

now

Platform Development Manager

Owns platform engineering for the company’s open-source infrastructure, and leads its AI-first shift: an Operations AI that investigates, diagnoses, and drafts the fix, with a person approving anything that changes a system.

leads

The team is the job

Most of my week is people work, and I run it on autonomy, mastery, and purpose: real ownership, work worth getting better at, and a reason it matters. I grow the individual to raise what the team can take on.

Cornell · Performance Leadership ’25
ask me

What am I like
to work with?

Ask how I decide, how I lead, what I have shipped. It answers from my verified profile.

Walk the timeline ↓
range

Many fields, one method

Mechanical and aerospace engineering, then software, SRE, operations, platform, and AI. Different domains, same method: define the problem, design the system, measure whether it worked.

The full track record ↓
MIT Professional Education · Applied Agentic AI ’26 U.S. patent application, turbine tuning ’17 AIAA conference paper ’13
portfolio

Products, not side projects

Personal builds I run end to end, on my own time. HurricaneWise repairs its own incidents. Longview runs my home on grounded data with human sign-off. The argument underneath both is written down.

Start with the argument → All build series →
ai-first · in production

HurricaneWise

A small product held to the practices a company would apply, run by one person. Detection is its permit: it attempts every failure its monitoring catches, proves the fix through the same checks any change of mine would face, ships it, and verifies the live site. I hold the envelope, the budget, and the switch. About $15 a month on mostly free tiers.

hurricanewise.com ↗ How it repairs itself → All 19 disciplines, one person →
reliability · incidents · observability · architecture · legal · branding and design · marketing and search · +12 more, every seat built and run with AI
Track record

Fifteen years, one through-line: make systems work.

  1. Platform Development Manager Feb 2024 – Present
    Vertex Inc.

    Owns platform engineering for the company’s open-source infrastructure, and leads its AI-first shift: an Operations AI that investigates, diagnoses, and drafts the fix, with a person approving anything that changes a system.

    • Sets technical strategy across Kubernetes, Terraform, event streaming, workflow infrastructure, and API gateway services.
    • Leads technical hiring for the platform engineering organization: writes the job descriptions, defines the engineering job functions, and designed the interview panels and stages with directors and VPs, the process behind more than 20 successful hires. He owns the final call on his own candidates, has built two teams that way (seven engineers and three interns, from Software Developer I through III plus a tech lead), and rebuilt the headcount plan through two reorganizations.
    • Scaling an Operations AI solution across the company with engineering, support, and operations partners, adoption still growing: services share one picture of how they depend on each other and how far a failure can spread, the AI diagnoses problems and drafts the fix, and a person approves anything that changes a system.
    • The same Operations AI writes the daily oncall handover, checking more than 60,000 datapoints (a number that keeps growing) against runbooks and checklists so the report says what actually happened, and it works incidents from root cause to the written report. Support agents get answers while the customer is still on the line, and the service catalog and the knowledge behind it stay current, release management included.
    • Designed and built a cross-department productivity dashboard, one shared view spanning customer success, service capabilities, self-service, operations, handover reports, and maturity frameworks.
    • Decides where AI belongs in the platform and where it does not, and trains the teams that have to run it.
    • Negotiates contracts and manages vendor relationships across the platform stack.
  2. Platform Development Manager Jan 2023 – Feb 2024
    NBCUniversal

    Owned the platform for NBC SportsNext, and moved it onto GitOps and infrastructure-as-code.

    • Led the transition to GitOps and Infrastructure-as-Code practices.
    • Set the platform roadmap for NBC SportsNext, and delivery and uptime improved under it, with the roadmap aligned across six engineering directors, cybersecurity, and IT.
    • Managed a team of six and about $1M in resources, and hired the team’s tech lead and operations lead.
  3. Senior Software Engineer → SRE Manager → Platform Development Manager Oct 2018 – Jan 2023
    Citrix

    Grew from senior IC to platform manager over ~4 years, leading reliability and platform strategy for Citrix Desktop-as-a-Service. Senior Software Engineer Oct 2018 to Mar 2020, SRE Manager Mar 2020 to Aug 2021, Platform Development Manager Aug 2021 to Jan 2023.

    • As Platform Development Manager, set platform strategy for Citrix Desktop-as-a-Service and shipped GA of the first fully hosted end-to-end DaaS offering on GCP with an eight-engineer team.
    • Ran a 12-member SRE team as SRE Manager through a consolidation: fewer tools, a leaner operation, and automation where the manual work was, which is where the gains came from. Operational toil dropped 35% and incident management improved, and he redefined SLAs, SLIs, and SLOs with automated dashboards behind them.
    • Led the cross-functional initiative that achieved SOC2 Type II certification, enabling key contract wins.
    • As Senior Software Engineer, architected an event-driven auto-remediation system that scanned and healed 30,000+ cloud nodes every 7 minutes, and served as acting manager for three core service teams (WEM, API Gateway, NGS).
    • Built the Splunk dashboards that carried insights, logs, security, and incident management across AWS, Azure, and GCP, and drove containerization (Docker) and self-healing systems.
    • Embedded platform components into application teams via inner-sourcing, reducing operational toil.
  4. Technical Lead May 2015 – Oct 2018
    Power Systems Mfg., LLC (Ansaldo Energia Group)

    The bridge from mechanical engineering into software.

    • Developed gas-turbine combustion software to automate real-time tuning for the PSM AutoTune product line, generating millions in annual revenue.
    • Established a Center of Excellence for cloud-based data streaming on AWS, ingesting nationwide engine data into a Monitoring & Diagnostics center.
  5. Engineer May 2013 – May 2015
    Parametric Solutions, Inc.

    Mechanical / turbomachinery engineering.

  6. Early engineering roles (aerospace & turbomachinery) 2010 – 2013
    Florida Atlantic University · Florida Turbine Technologies · Aerospace Technologies Group

    Research and test/QA engineering roots in aerospace and gas turbines, where the systems mindset started.

  7. Personal products Ongoing · own time
    Personal portfolio

    Practicing the craft end to end: mature, production-grade products built and run on personal time, for home and professional life, personal projects, not a commercial service.

    • HurricaneWise, an AI-first hurricane-tracking system designed, built, and operated solo: data pipeline, Claude-powered plain-language summaries, observability, incident automation, and infrastructure-as-code.
    • Longview proved its operating rules in daily use and is now maturing into its next generation, an early, ground-up rebuild that started with laws: an eight-article versioned constitution, an adversarial six-seat review board on every irreversible decision, physically separate per-household data, and roughly a dozen reserved powers only the human holds.
    • Goes deep across every layer a real platform has (frontend, backend, data, SRE, cost control, security, and CI/CD) while running almost entirely on free tiers.
    • Runs a live production website whose day-to-day operations, content, QA, SEO audits, security hardening, and backups, are handled by an orchestrated team of specialist AI agents working through the site’s own admin interface via browser automation.
    • Produces long-form strategy and research documents through structured multi-agent debate: a fact registry pinned first, specialist agents writing against it, and a contrarian reviewer that must approve the final output.

How I work

Build systems and teams that outgrow you.

I grow engineers past my own team, and build agent systems to carry the work, so a few people can own what used to take many.

Adoption that sticks beats demos that impress.

I prototype a lot and ship carefully. Demos are how I find the real requirement, so I build several throwaway ones to learn what people actually need, then hold the thing that ships to a higher bar: it has to survive real use, not a good meeting.

Automate the investigation, never the accountability.

The AI investigates, diagnoses, and drafts; a person owns the decisions that change a system. That line is what makes the autonomy safe to ship.

Honest feedback, early enough to change the outcome.

Careers are part of the job. I give the hard feedback while it can still change the result, and push for promotions before people have to ask.

Write the laws before the features.

My newest build began as eight versioned articles, before the first feature existed. Laws live in one place, preferences in another, and a compliance script checks every change against the whole tree.

Trust numbers you fetched, not numbers you remember.

Before planning my newest build, I re-verified the vendor limits it depends on against the vendors’ own published documentation, and the audit corrected my assumptions in several places. The same habit runs in public: HurricaneWise tracks its own spend and quota burn daily.

Point of view

Where agentic AI goes next.

Five opinions, labeled as opinions. I hold the first two strongly. The next two already run in systems I built, and the last one I have not tested yet.

Integration is the next moat.

Most of the world’s business still runs on things with no API: phone calls, emailed PDFs, aging portals. AI-written, human-reviewed integrations make connecting them affordable for the first time, and whoever connects them ends up holding the data everyone else is guessing at.

Small and mid-sized businesses have more to gain than their size suggests.

Small and mid-sized businesses feel every hire, so the specialist they cannot justify is capability they go without. Price has mostly stopped being the barrier. Implementation is, and that is the part AI is starting to absorb.

Agents end the dashboard era.

A dashboard waits to be asked. An agent notices on its own, briefs you, and drafts the follow-through for a human to approve.

A single agent flatters. A council catches.

One agent tends to agree with you. This site is reviewed by ten specialists who debate to alignment, with a contrarian who forces the hard case and a chair who has to rule. My newest build goes further: an adversarial review board attacks every irreversible decision before it ships.

AI is retiring the sunk cost of tool expertise.

Every managed tool used to charge an entry fee: days in its documentation, paid by a person and lost with that person, and that invested knowledge quietly locked stacks to vendors. AI pays that fee now. It reads the documentation, writes the configuration, reports the limits and the features you would not have known to ask about, and reviews the setup against what you actually need. So a managed tool should become a commodity, chosen on capability and price. I label that a hypothesis: I have not had to swap a vendor yet, and until I do it stays untested.

Some things need a human.

Ask-Felipe handles the basics, my background, my work, how I think. For anything that deserves a human, a question, an idea, a hello, leave a note and it reaches me directly. I read every one.