A PIT Crew for City Government
Can New York finally move beyond pilot purgatory?
New York City has never had a shortage of ideas.
Every year it pilots something: sensor networks, curbside composting, challenge-based heat-pump procurement in public housing, open streets. The city is very good at starting things. What it has never been good at is finishing them — carrying a pilot across the finish line from a successful experiment into a permanent public service.
That gap is not a technology problem. It is an institutional design problem.
On Monday, after a few laps around the Coney Island go-kart track, Mayor Mamdani launched the Public Interest Technology (PIT) Crew — five in-house teams of product managers, designers, engineers, researchers, and data experts, embedded across city agencies to build digital tools in a matter of months instead of years. With $5.24 million in baselined city funding, there will be thirty full-time hires, plus a fifth crew backed by the Rockefeller Foundation through the Mayor’s Fund. The first crew is already at work on a complaint portal, built with the Department of Consumer and Worker Protection, to enforce the city’s new Click to Cancel rule.
For years, many people, myself included, argued that the city should build permanent, in-house capacity to shepherd its own innovation. The PIT Crew is the closest thing yet, and its design directly targets three of the failures above.
Funding for PIT crew is baselined, meaning it is built into the ongoing operating budget, not funded through grant money or a one-off.
The PIT crew is also staffed as product teams, rather than consultants, embedded themselves from time to time inside agencies per project — building capacity directly within city government rather than renting it from vendors.
And the PIT crew is housed to persist, under OTI and reporting through Deputy Mayor for Operations Julia Kersen, making it a permanent group, not just a task force that will expire over time.
All of this is incredibly exciting, but as launched, the PIT crew still acts like a service-delivery team: it takes a known problem and ships the tool fast. Click to Cancel, for example, was promised by several politicians including ex-President Joe Biden.
But “building tools we already know we need” is different than addressing outstanding challenging problems that require innovative thinking and community input.
The PIT crew is created to de-risk execution, not innovation.
We don’t always know what the right tools are, and that requires a pilot to see whether a new innovation actually works in NYC. The entire point is to reduce that uncertainty at a small scale, allowing for the government to do R&D. In NYC today, a pilot can prove itself out, but with no defined next step, government moves on. The championing team leaves, the funding dries up, and the whole thing evaporates. A few years later a new team launches a new pilot solving the same problem from scratch.
The 2023 Pilot: NYC report, from NYCEDC and Cornell Tech, gave this problem a name: pilot purgatory. The city tests, learns, and then does nothing.
There are repeatable, systemic reasons for why pilots die:
Solution-first procurement: Agencies ask vendors to demonstrate technologies before defining the public need alongside residents. A pilot without a community-defined problem never develops the constituency it needs to survive a budget cycle or a change in administration.
No institutional ownership: When no permanent office is accountable for a pilot’s outcome, there is no one whose job it is to carry it forward. No one is owning the success of the project.
No legal pathway after validation: Even when a pilot works, the path of least resistance is to re-issue a full procurement cycle — a process so long and expensive that only funded departments and large firms can endure it.
Funding cliffs: Pilot funding is fought over and heavily reliant on grants and one-off allocations that expire on schedule regardless of results. The money runs out before the institution decides what to do.
Political turnover: Leadership changes, priorities reset, and initiatives tied to the last administration are left in the past.
Knowledge loss: When a pilot ends, its data and its learnings are often lost.
I’ve watched this play out many times before. For example, Numina ran validated pedestrian-sensor deployments across the city and never secured a contract; when the pilots ended, the data stayed siloed. OTI’s Smart City Testbed was paused when tech leadership changed — some analysis landed on City Share, but there was no formal home to preserve it. And earlier in my career, at a micromobility startup, we ran successful pilots and then watched an RFP that grew out of our own work stretch past eighteen months — with no clear communication and an incredibly delayed announcement.
I don’t have the full answer here, as people far closer to procurement law and agency operations than I am have wrestled with this for decades. But the PIT crew does have an opportunity to open up questions around the city’s innovation pipeline, and better connecting discovery, experimentation and scaling.
If the goal is to point this new capacity from the PIT crew upstream, here are a few ideas worth testing:
A problem-first intake: a lightweight community co-definition step that sets the problem and success metrics before anyone specs out a tool. Amsterdam runs a version of this in 90 days inside existing budgets.
Challenge-based procurement as the default approach — NYCHA’s Clean Heat for All already shows that this model works inside city government, pairing data-persistence and a defined scale-up criterion so success has somewhere to go.
A formalized research and university layer, strongly incorporating work and collaborating with groups like BetaNYC, Pilot City, universities, think tanks, and other research initiatives.
Success from the PIT crew shouldn't be only measured in tools shipped for solutions we already know. It should look to innovate and solve the right problems by testing more pilots and scaling successful ones into permanent programs; shorter procurement timelines; reusable digital infrastructure shared across agencies; institutional memory that survives political transitions; and community-defined problems that keep going after the pilot funding ends. The PIT crew succeeds if it unlocks the underlying systemic problems that restrict the city time and time again.
For twenty years, New York’s urban innovation problem was never the drivers or a shortage of ideas. There are countless startups, nonprofits, BIDs, and agency staff with good ideas and the willingness to test them. It was a failure of institutional design.
The PIT Crew may be the city’s best opportunity in a generation to change this — not because it builds software faster, but because the city can then begin to rebuilding the institutional machinery to turn experiments into real life public infrastructure across the city.