- Location
- Colombo
- Type
- Full-time
- Department
- Engineering
- Closing date
- Today
- Source
- CareersPage
Description
Job Description
The AI Integration Engineer builds the connective tissue between LLM APIs, existing product systems, and end users. This is not a model training role — it's about integrating, orchestrating, and deploying AI capabilities into production software, reliably, securely, and at scale.
Key Responsibilities
- Integrate LLM APIs (Anthropic Claude, OpenAI GPT-4o, AWS Bedrock, Google Gemini) into backend services and user-facing products
- Design and implement RAG pipelines: document ingestion, chunking strategy, vector store selection, retrieval tuning
- Build agentic workflows using frameworks such as AgentCore, LangChain, LlamaIndex, or custom orchestration patterns
- Manage prompt engineering, prompt versioning, and prompt evaluation frameworks
- Implement guardrails for LLM outputs: validation, content filtering, fallback logic
- Monitor AI system performance: latency, cost-per-query, accuracy drift, token usage
- Collaborate with frontend engineers to surface AI capabilities in product UIs
- Own the AI integration layer across the SDLC, from spec through CI/CD to production observability
Job Requirements
- 3+ years backend or full-stack experience; strong API design and consumption skills
- Proven experience integrating LLM APIs (any major provider) into production applications, not just prototypes
- Proficiency in Python and/or TypeScript/Node.js
- Hands-on experience with RAG: vector databases (Pinecone, Weaviate, pgvector, etc.), embedding models, chunking strategies
- Understanding of prompt engineering: system prompts, few-shot examples, chain-of-thought, structured output
- Solid grasp of API security, rate limiting, and cost management for LLM-based services
- Experience with AWS or another major cloud platform
Desirable Skills
- Experience with agentic frameworks: AgentCore, LangChain, LlamaIndex, CrewAI, AutoGen, or similar
- Familiarity with multi-modal AI (vision, audio) or function calling / tool use
- Understanding of fine-tuning workflows, even if not hands-on
- Experience using AI coding assistants to accelerate personal development workflow
Skills
PythonTypeScriptNode.jsAWSCI/CD
