AI Engineer

917Ventures; Philippines; On-site; PHP 1,200,000 - 2,000,000; Full time

Job summary

917Ventures in the Philippines is seeking an AI Engineer to build AI-powered products across its portfolio, integrating AI models with APIs and front-end interfaces.

You will design scalable prompts, orchestration, and CI/CD for AI features, working with LangChain, RAG, and foundation models to deliver production-ready solutions. Emphasis on reliability and usability in collaboration with product teams.

Qualifications

• Bachelor's degree in CS or related field.
• 3–5 years in software engineering, with 1–2 years in AI-enabled products.
• Hands-on experience deploying AI features in production.
• Experience with LLM APIs and building AI-powered applications.

Responsibilities

• Develop end-to-end AI-powered features across frontend and backend.
• Implement RAG architectures connecting LLMs to data sources.
• Design and maintain API layers (REST/GraphQL) for AI model inference and tool integrations.
• Collaborate with product teams to ship AI-native product experiences.

Skills

Front-end development

Back-end development

AI/ML

API design

DevOps

Education

Bachelor's degree in CS/SE/IT

Tools

LangChain

LlamaIndex

CrewAI

OpenAI SDK

Anthropic SDK

PostgreSQL

Vector DBs (Pinecone, Weaviate, Qdrant)

Job description

At 917Ventures, we build ventures that create lasting impact for the Globe Group ecosystem and beyond. As the corporate venture builder of Globe Telecom, we back bold ideas with the resources, data, and technology of one of the Philippines' largest digital platforms. We're looking for passionate builders who want to ship AI that matters.

Job Description

The AI Engineer is responsible for building both the \"brain\" (AI models, agents, and orchestration logic) and the \"body\" (APIs, integrations, and product surfaces) of AI-powered products across the 917Ventures portfolio. This role enables tight integration between application layers and AI capabilities — ensuring that AI features are not just functional but seamlessly woven into the user experience.

Operating across the full AI application stack, the AI Engineer bridges prompt engineering, model orchestration, and production deployment, with an emphasis on scalability, reliability, and usability. You will work hands-on with LLMs, agentic frameworks, and retrieval-augmented generation (RAG) architectures to deliver AI-native products from prototype to production.

Duties and Responsibilities

AI Application Development

• Build end-to-end AI-powered features and products — from responsive front-end interfaces to sophisticated AI back-ends, tool-use servers, and knowledge management systems.
• Implement RAG architectures, connecting LLMs to structured and unstructured data sources for grounded, context-aware responses.
• Design and maintain API layers (REST/GraphQL) that bridge front-end components with AI model inference, prompt pipelines, and external tool integrations.
• Take raw AI capabilities and turn them into polished, user-facing software products across portfolio ventures.

Agentic Systems & Orchestration

• Design and build agentic AI workflows, multi-step chains, and tool-use pipelines using frameworks such as LangChain, LlamaIndex, CrewAI, or Flowise.
• Engineer, version, and systematically evaluate prompts for LLM-based products; maintain prompt libraries and evaluation harnesses.
• Integrate foundation model APIs (OpenAI, Anthropic, Google, open-source) into applications with attention to cost, latency, and reliability.

DevOps & Lifecycle Management

• Develop and maintain CI/CD pipelines tailored for AI applications, ensuring smooth and automated deployment of models and services.
• Oversee best practices across the DevOps/MLOps lifecycle, including code repositories, API optimization, cloud resources, and data pipelines.
• Instrument AI systems with logging, tracing, and observability tooling to monitor model behavior in production.
• Collaborate with product managers, designers, and domain experts to define AI-native product experiences.
• Participate in cross-functional brainstorming, strategic planning, and business review sessions.
• Stay current with rapidly evolving AI capabilities and translate new developments into actionable product opportunities.

Nature of Problems Encountered

• Latency Management: Solving the “slow AI” problem — ensuring users aren't left waiting while a model processes a complex request, through streaming, caching, and architecture optimization.
• Prompt Engineering vs. UX: Aligning the hidden logic of AI prompts with the visible inputs and expectations of end users.
• State & Context Management: Handling the complex states of AI conversations — memory, context windows, history, and multi-turn reasoning.
• Non-Determinism: Debugging intermittent failures where an AI model provides different or incorrect outputs for the same input; building evaluation frameworks to catch regressions.
• Multi-Agent Coordination: Managing handoffs, shared state, and failure modes across agentic systems with multiple collaborating AI components.

KPIs

• Delivery velocity: number of AI features shipped from prototype to production per quarter.
• AI product reliability: uptime and error rates for AI-powered endpoints and agentic workflows.
• Retrieval quality: precision and recall of RAG pipelines as measured by evaluation harnesses.
• Functional AI Products: Production-ready applications where AI is a core feature — intelligent dashboards, AI-driven SaaS tools, conversational agents, and automation workflows.
• Integrated API & Agent Layers: Scalable middleware that handles prompt engineering, token management, tool-use routing, and model inference.
• Optimized Retrieval Systems: Efficiently indexed vector databases and retrieval pipelines for fast, accurate context retrieval in AI applications.
• Evaluation & Observability Pipelines: Automated testing harnesses, prompt regression suites, and production monitoring for deployed AI systems.
• Technical Documentation: Comprehensive guides for application architecture, API schemas, agent configurations, and deployment procedures.

Skills

Technical Stack

• Front-end: React.js, Next.js, or Vue.js; Tailwind CSS
• Back-end: Python (FastAPI/Flask) or Node.js (Express, Hono)
• AI/ML: LangChain, LlamaIndex, CrewAI, OpenAI SDK, Anthropic SDK; basic knowledge of PyTorch/TensorFlow
• Databases: PostgreSQL, and vector DBs such as Pinecone, Weaviate, Qdrant, pgvector, or Chroma
• Orchestration & Automation: Flowise, n8n, or equivalent workflow automation platforms
• DevOps: Git, GitHub Actions or GitLab CI, Docker, basic Terraform

AI & Agent Infrastructure

• Understanding of agentic patterns: tool use, multi-agent coordination, MCP (Model Context Protocol), and A2A communication.
• Experience with prompt engineering techniques, evaluation metrics, and systematic testing of LLM outputs.
• Familiarity with foundation model APIs and their trade-offs (GPT-4, Claude, Gemini, Llama, Mistral).
• Evidence of shipped AI-powered products or features, ideally showcasing LLM integration, RAG, or agentic systems.
• Contributions to open-source AI projects or a portfolio of AI experiments are highly beneficial.

Requirements

Education

• Bachelor's degree in Computer Science, Software Engineering, Information Technology, or related field (or equivalent practical experience).

Experience

• 3–5 years in software engineering or full-stack development, with at least 1–2 years specifically focused on AI-integrated products.
• Hands-on experience with LLM APIs and building AI-powered applications.
• Experience deploying AI features in production environments with real users.
• Strong communication and collaboration skills; ability to work cross-functionally in a fast-paced venture-building environment.

Nice to Have

• Experience building multi-agent systems or agent marketplace/discovery architectures.
• Background in fine-tuning or training models (LoRA, RLHF, DPO).
• Familiarity with the Philippine telco, fintech, or enterprise landscape.
• Experience in startup or venture-building environments with multiple concurrent products.