Job Overview
- Job Title: LLM / AI Engineer
- Function: Engineering – AI & Data
- Business Unit: REIL (Reliance Enterprise Intelligence Ltd)
- Location: Mumbai, India
- Experience: 5–8 Years
- Employment Type: Full-Time
- How to Apply: Send your resume directly to [email protected]
About REIL
Reliance Enterprise Intelligence Ltd (REIL) is a strategic joint venture between Reliance Industries and Meta. We combine Reliance’s unparalleled market scale and deep enterprise domain knowledge with Meta’s world-class AI and technology capabilities.
About the Programme
REIL is developing a cutting-edge enterprise AI platform focused on financial compliance and intelligence.
Role Overview
We are looking for an LLM / AI Engineer to build and anchor the language model layer of our platform. You will design and implement the systems that enable our platform to ingest compliance documents, accurately retrieve complex legal and regulatory knowledge, classify and triage corporate notices, and generate structured outputs (such as legal response drafts and recommendations).
Accuracy and reliability are non-negotiable in a compliance context. You will be directly responsible for ensuring that every LLM output is structurally grounded, explainable, and fully validated before it ever reaches an auditor or compliance officer.
Key Responsibilities1. Agentic AI & RAG Architecture
- Design and build the AI orchestration layer coordinating multi-step, automated compliance workflows.
- Build LLM-based document intelligence modules that extract, validate, and cross-reference structured data from financial documents to identify discrepancies and output fault reports.
- Architect and scale the Retrieval-Augmented Generation (RAG) backend that powers the platform’s legal and regulatory knowledge layer.
- Build data ingestion pipelines capable of parsing, chunking, embedding, and indexing complex legislation, circulars, tribunal rulings, and past compliance histories.
- Optimize the retrieval layer to surface high-relevance domain context using hybrid search, cross-encoder re-ranking, and dynamic context window management.
- Maintain and expand the core knowledge base to guarantee the system accurately reflects evolving legislation in real time.
2. Document Understanding & Classification
- Build enterprise-grade document understanding and semantic features for a variety of complex assets, including invoices, corporate notices, refund applications, and tax portals.
- Design and implement zero-shot/few-shot classification systems for incoming legal notices to dynamically extract key attributes, including issue type, relevant fiscal periods, amounts at stake, and statutory response deadlines.
3. Structured Output Generation
- Design robust prompt engineering frameworks that consistently generate structured, deterministic outputs (e.g., JSON schemas) ready for downstream execution.
- Build automated response-drafting systems for regulatory notices, ensuring every generated draft is rigorously grounded in the knowledge base with verifiable citation trails.
4. Evaluation & Quality Assurance
- Build a systematic, continuous evaluation framework to track LLM output parameters, specifically measuring accuracy, groundedness, citation truthfulness, and semantic consistency.
- Proactively mitigate hallucination risks by implementing alignment guardrails, confidence scores, and safe fallback behaviors for low-confidence inferences.
- Monitor production output quality closely to feed real-time performance data back into downstream prompt optimizations and retrieval training cycles.
Qualifications & SkillsEducation
- B.E. / B.Tech / M.Tech in Computer Science, Information Technology, Data Science, or a related quantitative field.
Required Experience
- 5+ years of core software engineering experience, with at least 3 years dedicated to building, deploying, and maintaining production-level LLM applications, NLP, or applied AI.
- Proven track record of deploying robust RAG architectures directly to enterprise production environments (not just sandbox prototypes).
- Strong experience working alongside deeply analytical or technical document corpora (legal, financial, tax, or regulatory texts).
Technical Skill Set
- LLM Orchestration: High proficiency with LangChain, LlamaIndex, or equivalent frameworks for managing multi-step agentic pipelines.
- Vector Infrastructures: Hands-on experience with Databricks Vector Search, Pinecone, Weaviate, or equivalent vector stores; proficiency executing hybrid keyword/vector searches.
- Prompt Optimization: Expertise in systematic prompt engineering, including chain-of-thought (CoT), few-shot tuning, and structured schema forcing.
- AI Evaluation Frameworks: Practical experience using tools like RAGAS or TruLens to benchmark retrieval quality and generation correctness.
- Document Processing: Heavy experience with PDF layout parsing, advanced OCR tools, and contextual chunking strategies.
- Programming & Libraries: Expert proficiency in Python and deep comfort navigating the Hugging Face ecosystem.
Preferred Qualifications
- Background in legaltech, tax tech, fintech, or highly regulated enterprise compliance landscapes.
- Familiarity with Indian regulatory documentation, such as GSTN notices, CBIC circulars, or appellate tribunal orders.
- Experience with target parameter fine-tuning or specialized domain adaptation of large language models.
- Understanding of multilingual document translation and semantic alignment over diverse regional Indian language inputs.
Core Competencies
- Technical: RAG Architecture & Knowledge Retrieval | LLM Evaluation & Output Quality | Prompt Engineering & Structured Outputs | Document Understanding & NLP | Production LLM Systems
- Behavioral: Precision & Attention to Detail | Cross-Functional Collaboration | Ownership & Accountability | Intellectual Curiosity | Structured Problem Solving
To Apply
Interested candidates who meet the above criteria are invited to submit their updated resume to [email protected] with the subject line "Application for LLM / AI Engineer - REIL".
Pay: ₹469,730.64 - ₹1,800,000.00 per year
Benefits:
- Health insurance
- Provident Fund
Work Location: In person
