NLP / LLM Engineer

Rubiscape; Pune District; On-site; INR 900,000 - 1,300,000; Full time

Benefits offered by this job

Structured output generation

Speech-to-text / Multimodal models

LLM safety & responsible AI

Open-source contributions

Job summary

Rubiscape, a Pune-based leader in decision intelligence, seeks an NLP/LLM Engineer to own the language stack for RubiAI and BusinessWorkbench. You will build pipelines, fine-tune open-source LLMs, and implement retrieval-augmented generation for enterprise analytics.

You will work with Python, LangChain, and PEFT tooling, collaborate with platform engineers, ensure latency SLOs, and contribute to multilingual capabilities in Indian languages.

Qualifications

• 3+ years of NLP engineering with production deployments.
• Deep proficiency with HuggingFace Transformers, Datasets, PEFT libraries.

Responsibilities

• Design and implement NLP pipelines for document understanding, entity extraction, classification, and summarisation.
• Fine-tune and evaluate open-source LLMs on domain corpora using PEFT techniques.
• Build and iterate on RAG architectures: chunking, embedding model selection, hybrid retrieval, re-ranking.
• Develop and maintain the prompt engineering framework for RubiAI: structured prompting, system prompts, regression testing.
• Evaluate LLM outputs using evaluation harnesses; track quality across versions.
• Implement multilingual NLP for Indian languages (Hindi, Marathi).
• Collaborate with AI Platform Engineer to deploy NLP microservices at scale with latency SLOs.

Skills

NLP engineering

HuggingFace

Python

RAG pipelines

LangChain

Vector stores

Model serving

LLM evaluation

Education

Bachelor’s or Master’s in CS/Computational Linguistics

Tools

Weaviate

Qdrant

Chroma

pgvector

LangChain

TorchServe

Job description

As an NLP / LLM Engineer at Rubiscape, youwill build the language intelligence that makes RubiAI — our agentic AI copilot— genuinely useful to enterprise analysts, data engineers, and businessdecision-makers interacting with complex data in natural language. You Will ownthe full NLP stack: from classical text processing and information extractionthrough LLM fine-tuning, retrieval-augmented generation, and prompt engineeringfor enterprise-grade accuracy and safety. This role directly advances Rubiscape’svision of a platform where every employee can query, analyse, and act onenterprise data without writing a single line of code.

Key Responsibilities

• Design andimplement NLP pipelines for document understanding, entity extraction, classification, and summarisation integrated into RubiAI and BusinessWorkbench.
• Fine-tuneand evaluate open-source LLMs (Llama 3, Mistral, Phi-3) on Rubiscape’sdomain-specific corpora using PEFT techniques (LoRA, QLoRA) and managedfine-tuning APIs.
• Build and iterate on RAG architectures: chunking strategies, embedding model selectionand evaluation, hybrid retrieval (dense + sparse), and re-ranking forhigh-precision enterprise knowledge retrieval.
• Developand maintain the prompt engineering framework for RubiAI: structured promptingpatterns, chain-of-thought templates, system prompt governance, and automatedregression testing for prompt quality.
• EvaluateLLM outputs systematically using RAGAS, DeepEval, or custom evaluationharnesses; track quality metrics across model versions and prompt iterations.
• Implementmultilingual NLP capabilities for Indian-language enterprise content (Hindi,Marathi) where required by government and public-sector customers.
• Collaboratewith the AI Platform Engineer to deploy NLP microservices at scale, ensuringlatency SLOs are met for interactive RubiAI copilot responses.

Nice to Have

• Experiencewith structured output generation (JSON mode, grammar-constrained decoding) forreliable LLM integration with enterprise data systems.
• Knowledgeof speech-to-text or multimodal models relevant to voice-driven analyticsinterfaces.
• Exposureto LLM safety, alignment, and responsible AI practices for regulated-sectordeployments.
• Contributionto open-source NLP or LLM tooling (HuggingFace, LangChain, LlamaIndex, orequivalent).

About Rubiscape

Rubiscape is India’s leading DecisionIntelligence Platform, unifying data engineering, BI, machine learning, andagentic AI in a single governed platform. Built in Pune and trusted by Fortune500 enterprises across BFSI, manufacturing, healthcare, and government. 8international innovation patents. 10 Industry-Academia Labs & COEs. From BIto AI — One Platform. Every Decision.

Requirements

• 3+ yearsof NLP engineering experience with demonstrable production deployments; atleast 1 year working with large language models in a product context.
• Deepproficiency with HuggingFace Transformers, Datasets, and PEFT libraries formodel fine-tuning and evaluation.
• Hands-onexperience with RAG pipeline construction using LangChain or LlamaIndex and atleast one vector store (Weaviate, Qdrant, Chroma, or pgvector).
• StrongPython skills and familiarity with model serving for NLP workloads (ONNX,TorchServe, vLLM, or Text Generation Inference).
• Solidunderstanding of LLM evaluation methodology: benchmark construction, humanpreference alignment, and hallucination mitigation strategies.
• Bachelor’sor Master’s degree in Computer Science, Computational Linguistics, or a relatedfield.