Artificial Intelligence Engineer

Aventra.AI; Gurugram District; On-site; INR 450,000 - 900,000; Full time

Job summary

Aventra.AI in Gurugram seeks an experienced AI/ML Engineer with strong hands-on expertise in Machine Learning, Python, PyTorch, LLMs, and RAG to join our team on-site in Gurugram.

We require 6+ years overall experience, 3+ years in ML, 4+ years Python, and mandatory RAG experience. Immediate joiners or those who can start within 0–15 days are preferred.

Qualifications

• 6+ years of overall experience.
• 3+ years of strong hands-on machine learning experience.
• 4+ years of strong Python programming experience.
• Strong hands-on experience with PyTorch.
• Strong practical experience with LLMs / Generative AI.
• Strong hands-on RAG experience is mandatory.
• Experience building production-ready RAG pipelines/applications.
• Strong understanding of embeddings, vector databases, semantic search, retrieval, and reranking.
• Experience with Transformers and LLM APIs.

Responsibilities

• Design, develop, and deploy ML and Deep Learning solutions for real-world business use cases.
• Build scalable AI/ML solutions using Python and PyTorch.
• Develop and implement LLM and Generative AI applications.
• Design, build, and optimize RAG pipelines.
• Work with embeddings, vector databases, semantic search, chunking, retrieval, reranking, and prompt engineering.
• Integrate LLMs with enterprise data and knowledge bases.
• Evaluate and optimize RAG and LLM applications for accuracy, relevance, latency, and scalability.
• Collaborate with engineering, product, and data teams to take AI solutions from POC to production.
• Write clean, maintainable, and production-grade Python code.

Skills

Machine Learning

Python

PyTorch

LLMs / Generative AI

RAG experience

Embeddings

Vector databases

Semantic search

Prompt engineering

Tools

LangChain

LlamaIndex

Hugging Face Transformers

Docker

APIs

MLOps

CI/CD

Job description

We are looking for an experienced AI/ML Engineer with strong hands‑on expertise in Machine Learning, Python, PyTorch, LLMs, and RAG to join our team in Gurugram.

RAG Experience: Strong hands‑on experience is mandatory

Notice Period: Immediate Joiners / Candidates who can join within 0–15 days

• Design, develop, and deploy Machine Learning and Deep Learning solutions for real-world business use cases.
• Build scalable AI/ML solutions using Python and PyTorch.
• Develop and implement LLM and Generative AI applications.
• Design, build, and optimize RAG (Retrieval-Augmented Generation) pipelines.
• Work with embeddings, vector databases, semantic search, chunking, retrieval, reranking, and prompt engineering.
• Integrate LLMs with enterprise data and knowledge bases.
• Evaluate and optimize RAG and LLM applications for accuracy, relevance, latency, and scalability.
• Collaborate with engineering, product, and data teams to take AI solutions from POC to production.
• Write clean, maintainable, and production-grade Python code.

• 6+ years of overall experience
• 3+ years of strong hands‑on Machine Learning experience
• 4+ years of strong Python programming experience
• Strong hands‑on experience with PyTorch
• Strong practical experience with LLMs / Generative AI
• Strong hands‑on RAG experience is mandatory
• Experience building production‑ready RAG pipelines/applications
• Strong understanding of embeddings, vector databases, semantic search, retrieval, and reranking
• Experience with Transformers and LLM APIs
• Strong understanding of ML/DL concepts, model evaluation, and optimization
• Excellent problem‑solving and coding skills

Good to Have

• Experience with LangChain / LlamaIndex
• Experience with Hugging Face Transformers
• Experience with LLM fine‑tuning / PEFT / LoRA
• Experience with AWS / Azure / GCP
• Knowledge of Docker, APIs, MLOps, CI/CD, and model deployment
• Experience working with open‑source LLMs

We are looking for someone who is hands‑on and strong in coding, with proven experience in Machine Learning, Python, and PyTorch, along with strong practical experience building LLM and RAG‑based solutions.