Founding Platform Engineer, Agent Runtime

Zingage; New York (NY); On-site; USD 180,000 - 250,000; Full time

Benefits offered by this job

Equity

Equipment stipend

NYC gym membership

Daily meals

Job summary

Zingage in New York invites a founding-platform lead to design and own the end-to-end agent runtime across four planes. The v1 is already funded and in flight, and you will shape the architecture, data flows, and governance that power every agent we ship.

You will join a small, dense team of engineers and operators, owning core decisions and driving the platform forward with hands-on implementation and strategic leadership.

Job description

Who we are

Home care is a $130B industry that runs on the telephone, and we are the AI that answers it. Zingage’s agents coordinate scheduling, on-call, and intake for 120+ home care agencies - 300K+ patient visits a month ride on our decisions, across every major EMR in the industry. We recently closed our Series A, backed by Bessemer, Bertelsmann Investments, and Yosemite (Reed Jobs), and the team is deliberately small and dense - operators and engineers from Ramp, Uber, Tennr, Datadog, and Verkada. Headquartered in New York.

The problem

Our agents act in the physical world: they fill shifts, move visits, and write to the system of record that governs care for frail patients. An agent is only as good as the runtime beneath it - and the runtime for production agents in a regulated industry does not exist yet. Building it means solving four problems your best colleagues would agree are open:

• Distributed systems where the component is the source of nondeterminism. Forty years of systems design assumes deterministic components in an unreliable world. An agent runtime inverts the assumption: the actor itself is stochastic. What does idempotency mean for an agent action? What is a transaction when one participant is an LLM and the other is a twenty-year-old EMR with no locking semantics?
• Freshness as a per-decision SLO.“How stale is too stale to act on?” has a different answer for a 2AM call-off than for next week’s schedule change. We need staleness budgets tied to real-world risk, enforced against four EMRs with no webhooks, brutal rate limits, undocumented semantics, and silent failure modes - CAP theorem where the partition is permanent and political. Those upstreams were never designed to be built on. That is not the obstacle; it is why this layer, once built, is the moat.
• Replay of a stochastic actor. You cannot replay the model, so you must replay the world: capture every agent trace - inputs, tool calls, upstream state - completely enough that a new model can be interviewed against two years of production reality overnight. Time-travel debugging for agents. The trace corpus this produces is the most valuable asset the company will ever own, because it is how we adopt every new model first while competitors spend a quarter stabilizing.
• Safety as unrepresentability. A production outage taught us the principle the hard way: no prompt can make the model emit a field the schema forbids. Your job is to generalize it - a capability and governance system where unsafe actions against a patient’s record are not discouraged or detected but unrepresentable at the boundary between a stochastic planner and a real-world effector. Zero unauthorized writes, audit-verified, forever, in a HIPAA-regulated industry where the audit trail must satisfy payers and regulators, not just engineers.

What winning means

For the world: autonomous care coordination that is auditable enough to be trusted with someone’s mother - 300K monthly visits that no longer depend on stale reads, unverifiable claims, or unsafe writes. For you: founding ownership of an entire layer. Not a slice of someone’s data platform - the whole runtime, all four planes, the architecture decisions, and in time the team that grows around it. The v1 of this system is already funded, owned, and in flight as our current engineering cycle; you are being hired to see what it actually is and build it into the platform every agent we ever ship runs on.

The build

Four planes, each wired to a number. The first three have their v1 in flight today under company-wide key results; the fourth is the horizon that makes this a five-year problem.

Plane

What it is The number

Data Polled and streamed EMR data across four vendors; per-decision freshness budgets; queryable on demand, never locked behind slow scrapes

≥99.9% of actionable reads inside their freshness budget; wrong actions caused by stale reads → 0

Action Every agent action traced, auditable, and replayable - the receipt ledger and the replay substrate Receipt coverage → 100%; then time-to-adopt a new model, measured in days

Governance Protected-field enforcement, capability-scoped writes, HIPAA-grade audit, payer-defensible records Unauthorized writes → 0, audit-verified, forever

Actuation The computer-use foundation: when an EMR has no API, the agent drives the screen - the universal adapter that ends our dependence on upstream permission Integration coverage without upstream cooperation

You build the first version of each plane, then hire the team that owns them. This is a founding-platform seat, not four jobs forever.

The scoreboard

Freshness SLO. Write-corruption count, where the only acceptable number is zero. Receipt and replay coverage. Time-to-adopt a new model. Engineering truth with your name on it, inside a culture that already runs blameless postmortems, ships risky changes behind flags with named rollback owners, and holds every fix to a 60-day durability standard before calling it done.

The offer

Top of market: a founding-platform grant, because we are hiring an owner, not a maintainer. New York, in person, at the founders’ table. If the phrase “the upstreams fight back” sounds like an attraction rather than a warning, we would like to meet you.

What we offer

• Competitive base and meaningful equity. If this works, it should work for you.
• Equipment stipend: whatever setup makes you fast.
• Luxury gym membership in NYC. We work in health care. Yours counts too.
• Daily lunch, and dinner when work runs past dark. Big fan of late-night jams!
• Time off as needed. We measure output, not hours at a desk.
• Happy hours, poker nights, and builder events in our Soho office.
• Snacks in office, including team favorites like Gruns gummies, sardines, protein bars, and midday froyo runs.

Interview process

1. Intro call — Connect with the Zingage founder or VP of Engineering. We’re looking for technical judgment, ownership mindset, and role clarity.
2. Live coding & system design — A real engineering problem, worked through live. We walk through the problem space together and ask you to design a system and reason through trade-offs in real time. No take-home — we move fast.
3. On-site — Hands-on coding and system design, similar to the live round but deeper. Panel debrief follows. Strong candidates receive an offer the same day.