About this opportunity:
We are hiring a GenAI Developer Intern (Postgraduate) to contribute to the development of production-grade Generative AI applications on AWS: LLM services, RAG pipelines, multi-agent systems, and AI-driven RAN optimization solutions. You will participate in lifecycle management for prompts, embeddings, models, agent tools, and knowledge bases, implement safety guardrails and observability, and support AI governance while collaborating across backend, DevOps, and product teams. If you are passionate about building scalable, secure GenAI platforms on AWS cloud and want to apply your skills in a telecom R&D environment, this role is for you.
What you will do:
- Contribute to design and delivery of production-grade GenAI applications (LLM services, RAG pipelines, multi-agent systems) on AWS
- Develop agentic AI workflows using LangChain, LangGraph, CrewAI, AutoGen, or OpenAI Agents SDK; including tool-calling, planning, reflection, and human-in-the-loop patterns
- Build and optimize Retrieval-Augmented Generation (RAG) pipelines: document ingestion, chunking strategies, embedding models, vector databases, hybrid search, and re-ranking
- Implement Model Context Protocol (MCP) integrations for standardized tool/resource exposure to LLMs
- Implement Agent-to-Agent (A2A) communication protocols for multi-agent orchestration
- Support lifecycle management (LCM) for GenAI components: prompts, policies, agent tools, knowledge bases, embeddings, and model versions; assist with versioning, rollbacks, and documentation
- Implement guardrails, safety layers, and output validation for secure, controlled LLM usage
- Build observability for GenAI workloads: structured logging, tracing, telemetry for prompt/response quality, retrieval metrics, and cost/latency dashboards
- Develop and deploy GenAI microservices on AWS (Bedrock, SageMaker, Lambda, ECS/Fargate, S3, OpenSearch Serverless)
- Write Infrastructure-as-Code (Terraform) for GenAI stack components on AWS
- Apply GenAI and LLM capabilities to RAN/telecom use cases: network optimization, fault prediction, configuration management, and knowledge extraction from 3GPP specifications
- Write clean, tested, production-quality Python code following software engineering best practices
- Participate in Agile ceremonies (sprint planning, daily standups, retrospectives) and code reviews
- Collaborate with cross-functional teams (backend, frontend, DevOps, data engineering, product) to deliver end-to-end AI features
The skills you bring:
Education
- Recent postgraduate (Master's or PhD) in Computer Science, Computer Engineering, AI/ML, or related engineering discipline, with a strong foundation in algorithms, data structures, software engineering, and mathematics (linear algebra, probability, statistics).
Generative AI & LLM Skills (Required)
- Hands-on experience building applications with Large Language Models, including use of orchestration frameworks such as LangChain, LangGraph, LlamaIndex, Semantic Kernel, AWS Strands, or AWS AgentCore.
- Strong understanding of RAG architectures, including chunking strategies (recursive, semantic, agentic), embedding models (OpenAI, Cohere, open-source), vector stores (Pinecone, Weaviate, Qdrant, ChromaDB, OpenSearch), and retrieval approaches (hybrid search, re-ranking, contextual compression).
- Experience with prompt and context engineering (system prompts, few-shot, chain-of-thought, ReAct), structured outputs (JSON mode, function calling, tool usage), guardrails, hooks, and skills.
- Familiarity with agentic AI design patterns, LLM evaluation methods (automated metrics, LLM-as-judge, human evaluation, benchmark design, MAS frameworks), and emerging protocols such as MCP (tool/resource integration) and A2A (multi-agent communication).
Software Engineering & Development Skills (Required)
- Strong Python proficiency (3.10+) including type hints, async/await, dataclasses, Pydantic v2, virtual environments, and package management (pip, poetry, uv).
- Experience building REST APIs (FastAPI/Flask), applying clean code principles, software design and architectural patterns (microservices, event-driven, API gateway), and version control (Git).
- Experience writing automated tests, working with CI/CD tools (GitHub Actions, GitLab CI, Jenkins), and understanding containerization (Docker) with basic orchestration concepts.
AWS Cloud & SaaS Skills (Required)
- Hands-on experience with AWS services such as EC2, S3, Lambda, ECS/Fargate, IAM, CloudWatch, API Gateway, and Step Functions, alongside AWS AI/ML tools (Bedrock, SageMaker, AWS Strands, AgentCore).
- Understanding of serverless and event-driven architectures, Infrastructure-as-Code (Terraform or AWS CDK), and cloud security best practices (IAM, least privilege, secrets management, VPC networking).
- Awareness of SaaS models, multi-tenancy patterns, and API-first design.
RAN & Telecom Domain (Preferred)
- Basic understanding of mobile network architecture (RAN, Core, Transport) and 4G/5G concepts, with willingness to quickly learn telecom domain.
- Interest in applying AI/ML to network optimization, fault management, and automation, with knowledge of O-RAN and RIC concepts considered a strong plus.
Additional Preferred Qualifications
- Experience with LLM fine-tuning (LoRA, QLoRA, RLHF, DPO), multimodal AI (vision/audio) and multimodal RAG, as well as GraphRAG and graph databases.
- Familiarity with observability tools (LangSmith, Langfuse, OpenTelemetry, Weights & Biases), data engineering, Agile/Scrum practices, and AI-assisted development workflows.
- Contributions such as research publications, open-source work, or technical blogging in AI/ML/GenAI.
- Knowledge of AI security practices including prompt injection mitigation, PII handling, and data governance.