Job Title: AI/ML Developer

Location: Pune, Maharashtra

Job Type: Full-time, Permanent

Company: Blue Polaris Inc.

Work Type: Hybrid

About Blue Polaris

At Blue Polaris, we are redefining how businesses make decisions. As a global leader in decision automation, AI, and business architecture, we partner with Fortune 500 companies to design intelligent, transparent systems using Decision Model and Notation (DMN) and our proprietary Decisions First™ approach.

We don’t just solve problems — we model how organizations think, decide, and act. By combining AI, analytics, business architecture, and human expertise, we help enterprises improve outcomes at scale.

As we expand into Latin America, we are building a hub in Panama and seeking top talent to shape the future of decision intelligence.

Key Responsibilities

• Design and implement AI/ML models using Python, with a focus on large language models (LLMs) such as GPT, LLaMA, and others.
• Integrate AI models into real-world applications using tools and frameworks like LangChain, WatsonX, Watson Studio, Amazon SageMaker, Azure Machine Learning, etc.
• Collaborate with cross-functional teams to ensure AI/ML models are well-integrated, scalable, and efficient within client solutions.
• Stay up to date with the latest AI research, model architectures, and tooling in both open-source and proprietary ecosystems.
• [Preferred] Implement or support fine-tuning of LLMs and vision models, applying best practices and efficiency techniques.

Preferred Additions to Your Skillset (Nice to Have)

• Experience with model fine-tuning techniques, including:
• Adapter-based methods (e.g., LoRA, QLoRA)
• Quantization (4-bit, 8-bit models)
• Parameter-efficient fine-tuning (PEFT)
• Hands-on familiarity with Hugging Face ecosystem, especially:
• Transformers
• PEFT
• Datasets
• Understanding of training fundamentals:
• Forward/backward passes
• Matrix factorization
• Gradients & optimization algorithms (e.g., AdamW)
• Overfitting control, rank selection, and hyperparameter tuning
• GPU/resource management experience, including:
• Training large models (7B+ parameters)
• Memory optimization across single or multi-GPU environments
• Familiarity with evaluation metrics like perplexity, BLEU, ROUGE, or task-specific benchmarks

Mandatory Skills & Qualifications

• Strong proficiency in Python
• 2–3 years of hands-on experience in machine learning projects
• Experience with LLMs and other deep learning architectures
• Experience with OpenAI APIs or other open-source foundation models
• Familiarity with tools like:
• LangChain, WatsonX, Watson Studio
• Azure ML, AWS SageMaker, Amazon Bedrock
• Git, Docker
• Proven ability to integrate statistical and AI-based models into applications

Education

• Bachelor’s degree in computer science, Engineering, or a related field (preferred)

Experience Requirements

• Total work experience: 2–3 years
• Python development: 2+ years
• Machine Learning: 2+ years
• LLMs or NLP-based systems: 1–2 years