Senior Backend Engineer - AI Agent Systems

Organization icon

Rize

Location icon

Jakarta, Indonesia

Apply now

? About Rize

Rize (rize.farm) is a leading AgTech startup making rice cultivation sustainable while improving farmer livelihoods across Asia. Rice farming is one of the largest sources of agricultural emissions globally—producing more methane than the entire aviation industry.

Backed by Temasek, Wavemaker Impact, Breakthrough Energy Ventures, and GenZero, we're building the technology infrastructure to help millions of smallholder farmers transition to climate-smart agriculture.

Current Impact:

  • 1,800+ farmers across Indonesia & Vietnam (targeting 10M+ by 2030)
  • 1,500 hectares under sustainable management (scaling to 55,000+ by 2027)
  • 35-45% methane emission reduction through Alternate Wetting & Drying (AWD) techniques

We've been featured in Bloomberg, Business Insider, and DealStreetAsia.

? The Opportunity: Build the Future of Agricultural Operations

We're at a critical inflection point—transitioning from human-directed operations to AI-assisted workflows. This role will build the foundation for autonomous agents that will eventually enable 10 million farmers to interact directly with our platform, without human intermediaries.

The Transformation:

  • Today: 50+ field agronomists manually schedule tasks, verify compliance, manage farmer relationships
  • Tomorrow: AI agents handle routine operations; humans focus on exceptions and relationship-building
  • End State: Platform scales to 10M farmers without proportional headcount growth


Your First Agent (Scheduling) Will:

  • Free up 50+ agronomists from manual task planning (currently takes 10+ hours/week per person)
  • Enable 5x scale (from 1,500 to 7,500+ hectares) without hiring proportional operations staff
  • Lay the groundwork for future agents: compliance review, farmer communication, supply chain optimization

This isn't "add AI to an existing system"—you're architecting the future operating model.


?️ What You'll Build

Current Tech Stack

  • Backend: Java, Spring Framework, Hibernate ORM, PostgreSQL
  • Frontend: React Native (mobile for agronomists), ReactJS (web console for managers)
  • Infrastructure: AWS (RDS, EC2), Kafka for event processing, Grafana/Prometheus monitoring
  • Integrations: WhatsApp API, Zoho Books (ERP), Xendit (payments), Mixpanel (analytics)


What You'll Own (First 90 Days)

Foundation

  • Design the Agent Control Plane: execution framework, decision logging, human override system
  • Set up LLM integration layer (OpenAI/Anthropic) with structured output validation
  • Build webhook listeners for operational events (e.g., Actual Planting Date changes)
  • Create audit trail system (every agent decision logged for learning and compliance)

MVP Delivery

  • Implement the first Agent with prompt-based decision making
  • Create Manager approval interface in Console (React/Ant Design)
  • Deploy to production with pilot Planting Groups

Production Hardening

  • Achieve 80% Manager approval rate without manual modification
  • Build feedback loop: human overrides → improve prompts → retrain decision logic
  • Implement monitoring: agent uptime, decision latency, override patterns
  • Design v2 roadmap: auto-approval thresholds, weather integration, multi-agent coordination

Beyond:

  • Expand agent platform
  • Integrate remote sensing data (satellite imagery for harvest prediction)
  • Scale to full autonomous mode (agents propose → humans approve occasionally → fully autonomous)


? What We're Looking For

Must-Haves:

  • 5+ years backend engineering (Java/Spring preferred; strong Python or Node.js engineers welcome)
  • Hands-on LLM application experience: Built production systems using OpenAI, Anthropic, or similar APIs
  • Prompt engineering expertise: Especially for structured outputs (JSON schemas, function calling, tool use)
  • Event-driven architecture: Webhooks, message queues (Kafka/RabbitMQ), async/background processing
  • Startup comfort: Ambiguity, speed, and ownership are part of the job. We move fast and ship iteratively.
  • Product mindset: You'll talk to Territory Managers, understand their workflows, and build for their needs


Nice-to-Haves:

  • Experience in AgTech, field operations software, or on-ground workforce management platforms
  • Geospatial data exposure: GPS, mapping, routing, or optimization problems
  • Multilingual LLM applications, especially Indonesian and/or Vietnamese
  • Experience with human-in-the-loop ML systems (agents that learn from human corrections)
  • Previous work in climate tech or social impact domains
  • Familiarity with computer vision for image validation (we'll build compliance agents next)


❌ You Might Not Be a Fit If...

  • You prefer working on well-defined specs with minimal ambiguity
  • You need extensive onboarding/hand-holding (we're lean and move fast)
  • You want to focus purely on coding (no ops calls, no talking to Territory Managers)
  • You're skeptical of LLMs in production (we're betting big on agentic systems)
  • You prefer backend-only work (you'll occasionally touch React for Console UIs)


You're a great fit if:

  • You're energized by building in the messy middle—where product requirements emerge from user feedback
  • You're comfortable shipping v1, learning from production usage, and iterating rapidly
  • You see AI agents as tools, not magic—you understand their failure modes and design for them
  • You want to build foundational infrastructure, not bolt-on features