Senior AI-Native Full-Stack Engineer

Knit; New York (NY); Hybrid; USD 200,000 - 275,000; Full time

Benefits offered by this job

Equity Options

Healthcare (medical, dental, vision)

Onboarding Technology Stipend

401(k) with company match

Flexible time-off

Hybrid working

Job summary

Knit in New York City is building an AI-native research platform and seeks exceptional engineers to shape core of the platform. As an early member, you will own 0→1 and 1→N features, build AI agent orchestration, and craft React/Next.js UIs.

You will ship fast with robust tests, instrument systems, and collaborate with Product, AI, Design, and customers. Expect high ownership, on-call duties, and improvements across latency, tooling, and enterprise-grade guardrails.

Qualifications

• Experience with LLM/agent systems, evals, and prompt tooling.
• Deep knowledge of event-driven systems, streaming, and workflow engines.
• Observability with OpenTelemetry/Prometheus/Datadog.
• Security/compliance background (SOC 2, GDPR).
• Bias to action: small PRs, daily deployments.
• In-person collaboration: NYC hub development; whiteboards above long threads.
• High ownership: touch product, talk to customers, influence roadmap.

Responsibilities

• Own 0→1 and 1→N features across the stack, from design docs to building high-throughput data pipelines and scalable FastAPI services.
• Architect and own the core agent orchestration engine, designing complex, multi-agent LLM workflows with guardrails and auditable actions.
• Build AI-native, human-in-the-loop UIs with React/Next.js for safe collaboration and transparency.
• Develop the experimentation platform: CI/CD for prompts and agents, evals, drift detection, safety layers, rollout controls.
• Ship fast, safely: instrument everything, write design docs, add tests, and build dashboards/alerts.
• Partner cross-functionally with Product, AI, Design, CS; talk to customers and SMEs; turn ambiguity into shipped product.
• Obsess over performance: latency optimization, caching, vector search, and infra efficiency.
• Operate what you build: on-call ownership and incident response.

Skills

LLM/agent systems

Event-driven systems

Observability

Security & compliance

Bias to action

NYC collaboration

High ownership

Tools

OpenTelemetry

Prometheus

Datadog

Job description

Knit in New York City is building an AI-native research platform and seeks exceptional engineers to shape core of the platform. As an early member, you will own 0→1 and 1→N features, build AI agent orchestration, and craft React/Next.js UIs.

You will ship fast with robust tests, instrument systems, and collaborate with Product, AI, Design, and customers. Expect high ownership, on-call duties, and improvements across latency, tooling, and enterprise-grade guardrails.