Generative AI Engineer

ABOUT VALIANCE

Valiance is a deeptech AI company building sovereign and mission-critical solutions for enterprises, public sector, and government institutions. From predictive maintenance and demand planning to sovereign AI for citizen services, we design systems that thrive in high-impact environments. Recognized with the NASSCOM AI GameChangers Award and Aegis Graham Bell Award, and as a Google Cloud Partner, our 150+ engineers and data scientists are shaping the future of industries and societies with responsible AI.

THE ROLE

We are looking for an AI Engineer who is not just proficient in using LLMs — but deeply understands how they work, knows when to apply them, and can architect and deploy production-grade Gen AI and Agentic AI systems. You will work on real-world deployments across Document Intelligence and Video Intelligence products, alongside solution delivery for enterprise and public sector clients.

This role is structured around a four-stage engineering capability ladder — from mastering LLM fundamentals to building MCP-integrated, tool-driven autonomous agent systems. You are expected to be competent across all four stages.

KEY RESPONSIBILITIES

LLM Engineering & Reliability

• Design and optimize prompts using zero-shot, few-shot, chain-of-thought, and role prompting techniques
• Manage context windows, token budgets, chat memory, and structured/JSON outputs effectively
• Implement validation, retry logic, hallucination handling, and safety guardrails in production pipelines
• Build FastAPI microservices that expose LLM capabilities as reliable, scalable APIs

RAG & Retrieval Engineering

• Architect full RAG pipelines — from tokenization and chunking through to output evaluation and traceability
• Work with vector databases, similarity search, and hybrid/ranking retrieval strategies
• Implement citation and provenance tracking to make AI responses auditable
• Select and tune embedding models appropriate to domain and deployment constraints

Agentic AI Systems

• Design multi-agent architectures with clearly defined roles, responsibilities, and boundaries
• Implement agent memory strategies — both short-term (in-context) and long-term (persistent)
• Build tool use and function calling pipelines; delegate and parallelize agent tasks effectively
• Develop and deploy agentic applications using LangGraph and related frameworks

MCP & Tool Integration

• Architect systems using the Model Context Protocol (MCP) — servers, clients, tools, and resources
• Enable RAG through MCP resources; design and implement multi-step prompt workflow templates
• Expose, integrate, and test tools for LLM-driven applications in agentic pipelines
• Build, test, and debug production-grade agentic applications using MCP end-to-end

WHAT YOU BRING

Technical Skills

• 4–6 years of hands-on software engineering experience; 2+ year in applied AI/LLM engineering
• Strong Python proficiency — you write clean, tested, production-quality code
• Working knowledge of the OpenAI APIs, LangChain/LangGraph, and at least one vector database (Pinecone, Weaviate, PGVector)
• Experience building and deploying REST APIs using FastAPI or equivalent
• Familiarity with Google Cloud Platform services (Vertex AI, Cloud Run, BigQuery preferred)
• Understanding of AI evaluation frameworks — you know how to measure and improve model output quality

Mindset & Behaviours

• You are deeply curious — you don't just use models, you understand what's happening inside them
• You take ownership end-to-end — from design to deployment to monitoring in production
• You can translate ambiguous client problems into structured AI solutions
• You communicate clearly with both technical peers and non-technical stakeholders
• You are comfortable with ambiguity and thrive in a high-paced, mission-driven environment