Job Title : AI/ML Developer Company : Blue Polaris (formerly Decision Management Solutions) Location : Pune District, Maharashtra, India (Hybrid) Employment Type : Full time Seniority : Mid-level (3-5 years experience)

About Blue Polaris

About the Role

We are looking for a mid-level AI/ML Developer to join our Pune team and contribute to the development of intelligent, production-ready AI systems. You'll work primarily on building and integrating AI models — LLMs, RAG pipelines, multi-agent architectures — while also contributing to the backend and frontend layers needed to bring those systems to life. This is a builder role; you'll be expected to take ownership of features end to end.

Key Responsibilities

- Build and optimise NLP pipelines, including fine-tuning, prompt engineering, and retrieval-augmented generation
- Develop and maintain backend APIs (FastAPI/Node.js) to expose AI models and connect them to frontend interfaces
- Work with vector databases (Qdrant, FAISS, Neo4j) and embedding models to support knowledge retrieval systems
- Contribute to MLOps practices including model versioning, monitoring, and deployment workflows
- Stay current with developments in GenAI, agentic AI, and open-source LLM tooling

Required Skills & Qualifications

- 3-5 years of hands-on software development experience, with at least 2 years focused on AI/ML
- Strong Python proficiency — you write clean, production-quality code
- Practical experience with LLMs and GenAI frameworks: LangChain, LangGraph, Hugging Face Transformers, or similar
- Experience building RAG systems or multi-agent AI pipelines
- Familiarity with NLP concepts: embeddings, token classification, text generation, semantic search




- Backend development experience with FastAPI, REST APIs, or Node.js
- Basic frontend capability (React.js or similar) sufficient to build interfaces for AI tools
- Experience with vector databases or graph databases (Qdrant, FAISS, Neo4j, or similar)
- Bachelor's degree in Computer Science, Information Technology, or a related field (or equivalent practical experience)

Nice to Have

- Experience with PyTorch — custom model training, fine-tuning (PEFT, LoRA, QLoRA), or ONNX optimisation
- Familiarity with IBM WatsonX, WatsonX Orchestrate, or WatsonX Assistant
- MLOps or LLMOps experience: model monitoring, CI/CD for ML pipelines, deployment tooling
- Exposure to cloud platforms: AWS (S3, EC2, SageMaker) or equivalent
- Experience with BPMN or workflow automation concepts

Pay: ₹1,200,000.00 - ₹1,400,000.00 per year

Benefits :

- Cell phone reimbursement
- Paid sick time
- Paid time off
- Provident Fund

Application Question(s):

- What is your current notice period as per your employment contract?
- What is your current CTC (Cost to Company) per annum in INR, and what are your salary expectations for this role?
- What is RAG (Retrieval-Augmented Generation) and have you built a RAG pipeline? If yes, what vector database or retrieval mechanism did you use and why?




- Describe your experience with Python for AI/ML development. What libraries and frameworks do you use regularly, and can you give an example of a complex problem you solved using Python?
- Have you worked with multi-agent AI systems or agentic frameworks such as LangGraph, CrewAI, or AutoGen? If yes, describe the architecture you built and the problem it solved.
- What is your experience with NLP tasks such as text classification, named entity recognition, semantic search, or language translation? Describe a specific NLP project you worked on.
- Have you worked with any frontend technologies such as React.js, Streamlit, or Gradio to build interfaces for AI systems? Describe what you built and its purpose.
- Describe your experience with model fine-tuning, prompt engineering, or any form of model customisation. What approach did you take and what was the outcome?
- How do you stay current with developments in AI and machine learning? Mention any recent development, paper, tool, or technique you have explored in the last 3-6 months.
- Describe a situation where an AI model or system you built did not perform as expected. What went wrong, how did you diagnose the issue, and what did you do to fix it?
- Are you currently based in Pune, Maharashtra? If not, are you open to relocating, and what would your timeline for relocation be?
- Do you have a GitHub profile, portfolio, or any publicly available work such as research papers, open-source contributions, or personal projects that you would like to share? If yes, please provide the links.

Work Location: Hybrid remote in Pune, Maharashtra (Pune, Pune District)

📌 Ai/ml Developer (Pune)
🏢 Decision Management Solutions
📍 Pune