Navajna Technologies - Generative AI Developer - LLM Models
Posted on: 11/07/2025
Job Description
Role Summary :
We are seeking an experienced Generative AI Developer to architect and build advanced AI-driven systems focused on document ingestion, semantic extraction, and decision support.
This role involves prototyping and productionizing fine-tuned or LoRA-augmented open-source language models as well as API-based large language models (LLMs).
The ideal candidate will be deeply familiar with LLM development, model fine-tuning, and system integration, with a strong emphasis on minimizing hallucinations, optimizing performance, and ensuring data sensitivity - especially within financial and climate domains.
Key Responsibilities :
- Architect, design, and build prompting and agent pipelines to ingest documents, perform semantic extraction, and provide intelligent decision support.
- Prototype, fine-tune, and productionize open-source LLMs such as Mistral, Qwen using LoRA or similar techniques, alongside API-driven models like GPT-4 and Claude.
- Develop and implement rigorous evaluation benchmarks focusing on model accuracy, latency, hallucination rates, and cost/token efficiency at scale.
- Collaborate cross-functionally with data engineering and backend teams to integrate real-time financial and climate data sources into AI workflows.
- Lead experimental research and development on retrieval-augmented generation (RAG), chain-of-thought prompting, and dynamic user profiling to enhance AI output relevance and personalization.
- Continuously optimize AI pipelines for production readiness, scalability, cost efficiency, and compliance with domain-specific constraints.
- Document technical designs, experimental results, and best practices for knowledge sharing and regulatory compliance.
- Stay abreast of emerging trends and advances in generative AI, LLM architectures, and related tooling.
Core Skills & Experience :
- 3+ years of hands-on experience developing with Large Language Models (LLMs) across platforms such as OpenAI, Anthropic, and Googles Vertex AI.
- Strong expertise in fine-tuning and customizing language models using LoRA, Hugging Face Transformers, or equivalent frameworks.
- Proven experience building and deploying agentic AI systems using LangChain or custom orchestration frameworks.
- Proficient in Python for model development, experimentation, and pipeline automation; familiarity with Java or JVM-based languages for system integration is a plus.
- Deep understanding of evaluation metrics and methodologies for generative AI models, with demonstrated ability to reduce hallucinations and improve output reliability.
- Knowledge of financial domain requirements and Environmental, Social, and Governance (ESG) or climate data sensitivity, explainability, and compliance considerations.
- Solid grasp of prompt engineering techniques including chain-of-thought, zero/few-shot prompting, and user profiling for context-aware generation.
- Experience integrating AI models with real-time data feeds, APIs, and complex backend systems.
Preferred Qualifications :
- Advanced degree (Masters or PhD) in Computer Science, AI, Machine Learning, or related field.
- Experience with cloud-based ML infrastructure (AWS Sagemaker, Google Cloud AI Platform, Azure ML).
- Familiarity with containerization (Docker, Kubernetes) and CI/CD pipelines for ML.
- Contributions to open source generative AI projects or publications in related fields.
- Experience with multi-modal AI systems or non-textual data integration
Did you find something suspicious?
Posted By
90
Posted in
AI/ML
Functional Area
ML / DL / AI Research
Job Code
1511385
