Campus-Trainee

Job Summary

The Platform Engineer will focus on building robust, production-ready solutions that solve real-world problems in intelligent automation, enterprise AI, and smart applications. This role is designed for strong Tier-1 fresh graduates (or 0-1 year experience) who combine deep knowledge of generative AI techniques (both foundational model fine-tuning and prompt engineering) with the software engineering skills needed to automate, optimize, and integrate LLM and multimodal AI pipelines into scalable platforms.

Key Responsibilities

Develop, integrate, and optimize end-to-end generative AI pipelines from data ingestion and preprocessing to model inference, output post-processing, and metadata generation.
Implement and fine-tune state-of-the-art LLMs and multimodal models for tasks such as text generation, summarization, retrieval-augmented generation (RAG), document understanding, code generation, and conversational AI.
Build and maintain automated training and evaluation pipelines (LLMOps/MLOps) that streamline model updates, benchmarking, prompt versioning, and deployment across cloud and edge environments.
Collaborate with platform engineers to optimize generative AI models for latency and throughput using tools like vLLM, TensorRT-LLM, ONNX Runtime, or quantization frameworks (GPTQ, AWQ, GGUF).
Create robust logic for complex AI workflow orchestration, including chaining multiple model calls, applying business rules, managing context windows, and filtering hallucinations or low-confidence outputs to generate reliable, production-grade responses.
Develop tools and scripts for dataset and prompt management, including automated data curation, quality checks, synthetic data generation, RLHF/preference data collection, and continuous evaluation loops.
Troubleshoot generative AI performance issues in diverse real-world conditions, including prompt sensitivity, latency spikes, and context limitations, and implement prompt engineering, fine-tuning, or architectural fixes.
Document AI pipeline architecture, model interfaces, evaluation benchmarks, and deployment guides for internal and external consumers.

Skill Requirements

Strong programming skills in Python (and optionally C++), with a focus on writing efficient, modular code for LLM and generative AI applications.
Solid understanding of Generative AI and NLP fundamentals, including tokenization, attention mechanisms, transformer architectures, embedding spaces, vector similarity, and prompt design principles.
Proficiency with core AI/ML libraries and frameworks, including HuggingFace Transformers, LangChain/LlamaIndex, NumPy, and at least one deep learning framework (PyTorch/TensorFlow) applied to generative tasks.
Hands-on experience through projects or internships developing and evaluating generative AI models, including LLM architectures (GPT, LLaMA, Mistral, Gemma), fine-tuning techniques (LoRA, QLoRA, instruction tuning), or RAG pipeline development.
Familiarity with data and inference concepts, including token limits and context window management, streaming inference, vector databases (Pinecone, Weaviate, FAISS, pgvector), and API-based model serving.
Understanding of software engineering basics, including version control (Git), unit testing, debugging, and integration of AI pipelines into larger applications.
Strong analytical skills to evaluate generative AI performance using BLEU, ROUGE, METEOR, BERTScore, hallucination rates, latency (TTFT/TPS), and resource usage (CPU/GPU/RAM/VRAM).

Other Requirements