AI Engineer

You will work closely with the Lead AI Engineer and Senior AI/ML Engineer, owning defined AI workstreams end to end. This role requires someone who can ship production features, troubleshoot real-world failures, and communicate clearly with technical leadership.

This is not a junior position. It is intended for an engineer who is technically one level below Senior, but operates like one - takes ownership, writes maintainable code, debugs independently, and knows when to escalate versus when to solve.

If most of your AI experience has been limited to proofs of concept, notebooks, or internal-only tools, this role will likely be a stretch at this stage.

What You Will Work On

LLM Integration & Prompt Engineering

• Build and maintain LLM-powered features using OpenAI, Anthropic Claude, or Gemini APIs
• Design prompts that perform reliably in production environments, not just prototypes
• Implement RAG pipelines, function calling, and structured output parsing
• Manage prompt versioning, failure handling, and cost-performance trade-offs

Conversational AI

• Build and maintain chatbot and AI assistant flows for WhatsApp, web chat, and in-app experiences
• Handle multi-turn conversations, context persistence, intent transitions, and escalation logic
• Integrate conversational AI with client backend systems and enterprise workflows

Document Intelligence

• Integrate and support OCR-driven document processing pipelines
• Extract structured data from real-world documents such as invoices, KYC forms, and insurance policies
• Route validated outputs into downstream systems for automation and analytics

AI Quality, Evaluation & Monitoring

• Monitor production AI systems for accuracy drift, response quality, and latency
• Implement logging, evaluation workflows, and quality checks
• Identify failure patterns and work with senior engineers to improve prompts, pipelines, or models

Backend Integration Support

• Support backend integrations between AI services and enterprise APIs
• Build Python-based microservices, Lambda functions, and API wrappers
• Ensure AI components are secure, observable, and production-ready

What We're Looking For

Core Requirements

• 4-5 years of professional software engineering experience, with at least 2 years focused on AI/ML or LLM systems in production
• Strong Python skills - able to build and maintain production services (not just scripts), including:
  • Error handling
  • Logging
  • Testing
• Hands-on experience working with LLM APIs (OpenAI, Anthropic, or Gemini) beyond tutorials
  • Understands token limits, latency, cost management, prompt iteration, and failure modes
• Experience implementing Retrieval-Augmented Generation (RAG) in a production or near-production environment
• Experience building multi-turn conversational AI systems that handled real user traffic
• Comfortable deploying Python services on AWS (Lambda or EC2)
  • Not an infrastructure specialist, but capable in a cloud environment
• Highly self-directed - able to take requirements, ask the right questions, and deliver without micromanagement

Nice to Have

• Experience with LangChain or LlamaIndex beyond basic examples
• Production experience with Document AI tools (AWS Textract, Azure Form Recognizer, AllXtract)
• Voice AI experience (STT/TTS) using tools such as Deepgram, Whisper, or ElevenLabs
• Experience building AI services using FastAPI or Flask
• Hands-on use of Hugging Face (model evaluation, fine-tuning, or inference pipelines)
• Exposure to Southeast Asian enterprise clients (Malaysia, Indonesia, regional markets)
• Proficiency in Mandarin, Malay, or Bahasa Indonesia