AI Engineer

Unosecur; Bengaluru; On-site; INR 4,000,000 - 7,000,000; Full time

Job summary

Unosecur is seeking a senior AI Engineer to own GenAI and agentic architectures in Bengaluru, driving LLM-based solutions and secure AI capabilities. You’ll lead end-to-end design, mentor engineers, and collaborate with product and security teams to deliver scalable AI features on AWS and GCP.

You will leverage LangChain, LangGraph, MCP, and RAG pipelines to build robust, production-ready AI services that integrate with our security SaaS platform, while navigating startup-pace challenges and

Qualifications

• 10+ years of overall engineering experience.
• 3+ years hands-on Generative AI / LLM-based solutions.
• Experience designing and architecting AI solutions.
• Depth in agentic / GenAI stack: LangChain / LangGraph, MCP, RAG, vector databases, and multiple LLMs / SLMs.
• Working knowledge of GenAI platforms (AWS Bedrock, Vertex AI) and cloud (AWS / GCP).
• Strong fundamentals: data structures, problem solving, scalable distributed services.

Responsibilities

• Design and architect end-to-end GenAI and agentic systems—from problem framing through orchestration, retrieval, and deployment.
• Build agentic workflows using LangChain / LangGraph and MCP, integrating tools, plugins, and data sources with clear boundaries.
• Stand up and improve RAG pipelines and vector/embedding layers.
• Own model-selection strategy across LLMs and SLMs, balancing quality, latency, cost, and security.
• Raise engineering bar for reliability, observability, and quality of AI services in production.
• Partner with product and security to translate ambiguous problems into safe AI capabilities.
• Mentor engineers and act as the technical reference point for GenAI.

Skills

Generative AI

LangChain

LangGraph

MCP

RAG pipelines

Vector databases

LLMs/SLMs

AWS Bedrock

Vertex AI

Cloud AWS/GCP

Job description

AI Engineer ? Senior AI Engineer / GenAI Engineer Internal level: SDE-3 (senior individual contributor; calibrated to IC4) Job family: Software Engineering / Machine Learning Experience: 10+ years overall (2+ years hands-on Generative AI)

Company summary Founded by Santhosh Jayaprakash and headquartered in Berlin, Unosecur is a fast-growing, Pre-Series A startup with $5M in funding. We're a fast-growing B2B security SaaS platform making identity security smarter and simpler for enterprises worldwide. You?ll be part of a diverse team that thrives on creativity, collaboration, and cross-border problem-solving. With cybersecurity now mission-critical, you?ll be building not just a career, but a future in one of tech?s most dynamic and resilient sectors.

What your experience will be You will sit on the engineering team and own the design and build of our Generative AI and agentic capabilities ? the LLM-backed services, agent orchestration, and the architecture that connects them to our security platform. You?ll work closely with the founding and senior engineering team and report into engineering leadership, partnering day to day with backend, cloud, and product. Your toolkit spans LLMs and small language models, agent frameworks such as LangChain and LangGraph, MCP, RAG pipelines and vector databases, and GenAI platforms like AWS Bedrock and Google Vertex AI, deployed on cloud infrastructure (AWS / GCP). This is a Bengaluru-based role.

Why you belong here and how you will grow This is a small, senior team where the person who designs a system is the person who ships it ? you won?t be handed someone else?s blueprint. You?ll set the technical direction for AI in the product, mentor engineers around you, and build credibility with founders who are close to the work. Because we are early, the surface area is wide: you?ll grow as an architect, as a hands-on builder at the edge of the GenAI field, and as a technical leader whose decisions visibly move the company.

What success looks like

• Design and architect end-to-end GenAI and agentic systems ? from problem framing through orchestration, retrieval, and production deployment.
• Build agentic workflows using LangChain / LangGraph and MCP, integrating tools, plugins, and external data sources with clear permission and failure boundaries.
• Stand up and continuously improve RAG pipelines, including retrieval quality evaluation and the vector / embedding database layer underneath them.
• Own model-selection strategy across LLMs and SLMs, balancing quality, latency, cost, and the security and data-handling constraints of our domain.
• Raise the engineering bar ? reliability, observability, and quality ? for AI services running at production load.
• Partner with product and security to translate ambiguous problems into well-architected, safe AI capabilities.
• Mentor engineers and act as the technical reference point for GenAI across the team.

Requirements

• 10+ years of overall engineering experience, including 3+ years hands-on in Generative AI / LLM-based solutions.
• Demonstrated experience designing and architecting AI solutions ? not just implementing or consuming AI tools. You can point to systems you personally built.
• Depth in the agentic / GenAI stack: agent orchestration (LangChain / LangGraph), MCP, RAG, vector databases, and multiple LLMs / SLMs with a real point of view on model selection.
• Strong engineering fundamentals ? data structures, problem-solving, and the ability to build scalable, reliable distributed services.
• Working knowledge of a GenAI platform (AWS Bedrock, Google Vertex AI, or similar) and cloud (AWS / GCP), plus cloud security fundamentals.
• Behavioral: a builder’s bias for ownership, comfort operating with startup ambiguity, and the judgment to make sound architecture trade-offs.

Nice to have

• Open-source AI contributions or substantial open-source hands-on use.
• Experience building AI into a security, identity, or other compliance-sensitive product.
• RAG evaluation tooling (e.g., Ragas) and structured approaches to measuring AI quality.
• Product-based or startup background where you owned systems end to end.

Benefits