Full-stack AI Engineer

Full-stack AI Engineer

Nova Platform · GoComet

Team

Nova — Autonomous Outcome Engineering (AOE)

Experience

0–4 years (2024 / 2025 / 2026 graduates)

Openings

12

The short version

We're building Nova — a governed AI platform that rewires how enterprise logistics actually works. Not dashboards. Not chatbots. Real agentic workflows that extract documents, validate data, orchestrate multi-step approvals, and make decisions — all running on structured knowledge, not vibes.

You'll write production code, sit in front of enterprise clients, and ship AI-powered workflows from day one. If you want a role where you build and think and ship — not just one of the three — keep reading.

What you'll actually do

Build real things

You'll write production code across the full stack — React frontends, Python/Node backends, ClickHouse queries, Kafka event pipelines, LangGraph agent workflows, and YAML-driven process definitions. Not prototypes. Not demos. Stuff that runs in production for real companies moving real cargo across real oceans.

Talk to enterprise clients

You'll sit across from logistics managers, freight forwarders, and operations leads. You'll listen to their messy, exception-filled processes — 6 approval levels, 4 document types, 3 systems that don't talk to each other — and turn that chaos into structured workflows they can actually use. This isn't optional. It's the job.

Think AI-first

Every problem you encounter, your first question should be: Can an agent handle this Not let me build a CRUD app. You'll design agents that extract fields from Bills of Lading, validate invoices against purchase orders, auto-route approvals based on business rules, and flag exceptions before humans even notice them. You'll work with LLMs, RAG pipelines, document parsers, and agent orchestration frameworks daily.

Own problems end-to-end

You won't hand off a spec and wait. You'll identify the problem, design the solution, build it, configure it for the client, deploy it, watch it in production, and fix what breaks. The FDE model means you own outcomes, not tasks.

What Nova actually is

Nova is GoComet's platform transformation. Four pillars:

Workflow Orchestrator

Configurable, visual, multi-step business processes. Clients define their own approval chains, routing logic, and automation triggers through a drag-and-drop graph editor. YAML under the hood. React Flow on the screen. Zero hardcoded business logic — the engine is completely generic.

Agents Orchestrator

AI agents that handle analytics, document extraction, validation, monitoring, and recommendations. Built on LangGraph with a five-stage pipeline:

scope resolution → context compilation → schema routing → plan + execute → evidence delivery

Every agent runs with governed business context, not ad-hoc prompting.

No-Code App Builder

Custom micro-apps attached to workflow steps. When a workflow node needs more than Approve/Reject — a rate comparison panel, a document review interface, an exception dashboard — it's built here.

Data Layer

Unified ingestion from GoComet products, external sources, and client uploads. ClickHouse as the analytical store. Everything tagged, mapped, and queryable. Tenant-isolated by construction.

The tech you'll work with

Frontend: React, React Flow (node-based graph editors)

Backend: Python, Node.js, FastAPI

Data:

• ClickHouse
• Kafka (event streaming)
• Debezium (CDC)

AI/ML:

• LangGraph (agent orchestration)
• LLMs (prompt engineering, RAG, tool use)
• DPT-2 (document extraction)
• PageIndex (long-document reasoning)
• Weaviate (vector search)

Permissions: OpenFGA (Zanzibar-based relationship access control)

Observability: Langfuse (LLM tracing), OpenTelemetry

Cost control: LiteLLM (gateway + budgets), Orkestra (model routing)

Infrastructure: dbt (semantic layer / MetricFlow), Temporal (scheduling), DataHub / OpenMetadata (catalog)

You don't need to know all of this on day one. You need to be the kind of person who can pick up any of it in a week.

What we're looking for

The non-negotiables

• You can code. Full-stack, independently, at production quality.
• You've built things — not just assignments, but projects, tools, systems that other people actually used.
• You're comfortable across the stack and don't need hand-holding to ship a feature.
• You can communicate technical concepts to non-technical clients.
• You can run discovery calls, write clearly, and present effectively.
• You think in systems and question complexity.
• You learn fast and adapt to new domains quickly.

Strong signals

• You've built something with LLMs — not just used ChatGPT, but integrated it into a working system
• You've worked on a side project or open-source contribution outside core CS
• You can structure ambiguous problems into clear plans
• You default to shipping over perfecting
• You've done client-facing or consulting-style work

What this role is NOT

• Not a pure backend/frontend engineering job
• Not a PM-only role
• Not a consulting-only role
• Not a support/configuration role

Why this is different from your other offers

• You won't be a cog — you'll own an entire client's AI workflow end-to-end
• You'll work on AI that actually makes decisions in enterprise environments
• You'll learn both engineering and business deeply
• You'll ship from week one
• You'll join a founding team building a major platform
• You'll work directly with leadership in a 2-in-a-box model

About GoComet

GoComet is a supply chain visibility and automation platform used by enterprises across 30+ countries.

We track shipments across ocean, air, road, and rail. We manage procurement, contracts, invoices, and compliance. And now, with Nova, we're building the AI-native operating layer that ties it all together.

Logistics moves $12 trillion in goods every year. Most of it still runs on spreadsheets, email chains, and phone calls. We're fixing that — not with another dashboard, but with intelligent, governed, agent-driven automation.