About Xeo Software
Xeo Software is an AI-first company with a 12-member high-performance team building enterprise-grade AI systems for internal automation and client solutions. We focus on practical, scalable AI systems and not prototypes.
We are looking for a RAG & AI Agent Engineer who will take ownership of retrieval systems and AI agent architectures the same way a Database Administrator owns data infrastructure.
Our Core Values Come First
At Xeo Software, we believe skills can be taught, but values must be shared. To thrive here, you must not only be technically strong you must live out the principles that define how we work together and with our clients.
Two-Way Candor
We speak directly and respectfully, even when conversations are hard. We invite honest feedback and give it in return, knowing it strengthens trust and outcomes.
Excellence & Professionalism
We take pride in doing things right the first time. That means going beyond good enough polishing details, documenting solutions, and holding ourselves to the highest standard.
Continuous Learning
Technology evolves fast, and so do we. We actively seek out new tools, share knowledge with teammates, and integrate innovations that drive better results.
Loyalty to Team & Clients
We succeed together. We support one another, assume good intent, and show the same loyalty to our clients that we do to our coworkers.
If these values resonate with you, you'll feel at home here. If not, no matter your technical skills, this isn't the right place for you.
Role Overview
This role is for someone who understands that retrieval quality is infrastructure.
You will design, build, optimize, evaluate, and maintain Retrieval-Augmented Generation (RAG) systems and AI agents from scratch using retrieval-based approaches.
You will work across:
- Internal AI automation systems
- Client-facing AI deployments
- Production-grade LLM applications
This is a system ownership role, not just feature development.
Key Responsibilities
RAG System Architecture
- Design and build RAG systems from scratch
- Implement optimal chunking strategies and embedding pipelines
- Architect vector database schemas and indexing strategies
- Optimize retrieval latency and accuracy
- Improve context injection strategies for LLMs
AI Agent Engineering
- Build production-grade AI agents using Python & LangChain
- Develop tool-integrated agents (APIs, databases, workflow engines)
- Implement multi-step reasoning and orchestration logic
- Improve agent reliability and eliminate hallucinations
Infrastructure Ownership (AI-DBA Mindset)
- Maintain and monitor vector database performance
- Diagnose retrieval failures and context misalignment
- Ensure scalability, logging, observability, and cost optimization
- Continuously tune embeddings and retrieval pipelines
Evaluation & Quality Control
- Build RAG evaluation pipelines
- Implement hallucination detection mechanisms
- Measure retrieval precision/recall and answer grounding
- Design benchmarking frameworks for client deployments
Production Deployment
- Deploy production-grade LLM systems
- Ensure robustness, security, and scalability
- Work closely with product and engineering teams
- Adapt architectures for different client use cases
Required Skills
- 36 years of experience in software development
- Strong proficiency in Python
- Hands-on experience with:
- LangChain
- Vector databases
- OpenAI APIs
- Proven experience building retrieval-based RAG systems
- Strong understanding of:
- Embeddings
- Chunking strategies
- Indexing optimization
- Context window management
- Experience deploying LLM systems in production environments
Preferred
- Prior production RAG experience (strongly preferred)
- Experience optimizing vector search performance
- Experience building evaluation pipelines for LLM systems
- Exposure to hallucination mitigation techniques
- Experience working on client-facing AI projects
What We're Looking For
- Someone who thinks like an AI Infrastructure Owner
- Strong debugging and system-thinking ability
- Ability to diagnose failures methodically
- Comfortable working in a small, fast-moving AI-first team
- Strong problem-solving mindset
Why Join Xeo Software
- Be part of a focused 12-member AI-first team
- Build production AI systems and not experiments
- High ownership and architectural impact
- Opportunity to shape AI infrastructure at company scale
- Transition from remote to in-office collaboration in Ahmedabad
