Associate AI/ML Engineer: Generative AI Specialist

Narba Consulting; Dadri; Hybrid; INR 300,000 - 420,000; Full time

Job summary

Narba Consulting in Noida invites a motivated Associate AI/ML Engineer to join our Generative AI team. This entry-level role focuses on turning ideas into production-ready Python services using FastAPI or Flask, with emphasis on practical AI systems and reliable APIs.

You will read docs, debug integration issues, write clean code, and work on latency, context limits, and output quality. If you have 1–2 projects or internships and a graduate degree, you fit well.

Qualifications

• Proficiency in Python and solid CS fundamentals.
• Experience reading docs, debugging API integrations, and writing clean code.
• Familiarity with building production-grade AI workflows and latency considerations.
• Graduation with 0–1 year experience or internship in software/ML.

Responsibilities

• Build and maintain back-end services that consume model APIs using Python (FastAPI/Flask).
• Implement basic RAG pipelines using vector databases and chunking strategies.
• Develop prompts and evaluate output quality with structured formats (JSON).
• Create quick demo interfaces using Streamlit or Gradio for internal testing.
• Write unit tests and help optimize latency and token usage.

Skills

Python

OOP concepts

Data structures

RESTful APIs

Asynchronous programming

Problem-solving

Technical projects/ internships

Education

Bachelor's degree

Tools

FastAPI

Flask

Streamlit

Gradio

ChromaDB

FAISS

LangChain

LlamaIndex

Hugging Face

Docker

AWS

Job description

Associate AI/ML Engineer: Generative AI Specialist About the Role

Generative AI is moving fast, but taking models from a Jupyter Notebook into production is where the real work begins. We are looking for an Associate AI/ML Engineer who has solid fundamentals in computer science, strong Python skills, and a genuine interest in building practical AI systems.

This is an entry-level technical role. You won't be expected to invent new architectures on Day 1, but you will be expected to read documentation, debug API integration issues, write clean code, and learn how to manage LLM latency, context limits, and output quality.

Position Details

• Role Type: Full-time
• Experience Required: 0--1 Year
• Location: (Hybrid) Noida, India
• Eligibility: Graduate

Key Responsibilities

• API Integration & App Development: Build and maintain back-end services that consume OpenAI, Gemini, or open-source model APIs using Python (FastAPI / Flask).
• RAG & Vector Search: Implement basic Retrieval-Augmented Generation (RAG) pipelines using vector databases (e.g., ChromaDB, FAISS) and chunking strategies to query custom datasets.
• Prompt Engineering & Evaluation: Test, iterate, and benchmark prompts to reduce hallucinations and enforce structured output formats (e.g., JSON schemas).
• Prototyping: Build quick, functional demo interfaces using Streamlit , Gradio, or simple front-end setups for internal testing.
• Code Maintenance & Debugging: Write unit tests, handle API rate limits/errors gracefully, and assist in optimizing latency and token usage.

What We Expect From You

• Strong Programming Core: Proficiency in Python (understanding OOP concepts, data structures, and standard libraries).
• Database & SQL Basics: Good command over relational databases (SQL joins, queries) and familiarity with vector embeddings.
• API & Web Fundamentals: Basic understanding of RESTful APIs, HTTP requests, and asynchronous programming in Python.
• Problem-Solving Ability: Capacity to read complex technical documentation, debug broken pipelines, and work through logical problems independently.
• Demonstrated Interest: At least 1--2 technical projects, internships, or open-source contributions involving software development or machine learning.

Bonus Points

• Experience with frameworks like LangChain , LlamaIndex , or Hugging Face .
• Basic exposure to cloud deployments (AWS / Azure / GCP ) or Docker .
• Participation in coding platforms, technical hackathons, or open-source projects.