Solutions Architect - Backend (Agentic AI & Python)

EngiNeo Solutions; Gurugram District; On-site; INR 1,500,000 - 2,100,000; Full time

Job summary

EngiNeo Solutions seeks a senior backend engineer to design and scale agentic AI systems, orchestrating multi-step tool calls and live streaming of LLM responses using FastAPI. You will build robust backends, implement RAG patterns, and ensure reliability with secure authentication and scalable architectures.

You will mentor teams, participate in agile ceremonies, and collaborate with product and data teams to deliver production-grade APIs on AWS, GCP, or Azure with Docker/Kubernetes.

Qualifications

• Bachelor's or Master's degree in Computer Science, Information Technology, Electronics, or a related engineering discipline.
• 7-10 years of hands-on backend engineering experience, with a strong focus on building and scaling production services and APIs.
• Proven, production-grade experience designing and building agentic AI / LLM-orchestration workflows using frameworks such as LangGraph, AutoGen, or CrewAI, or robust custom tool-calling implementations.
• Demonstrated ability to instruct an LLM to reliably reason, trigger external function calls, and manage conversational state and memory in production.
• Expert-level proficiency in Python, including asynchronous programming, with deep experience in FastAPI.
• Strong expertise in real-time streaming (SSE / WebSockets) for LLM responses and asynchronous/event-driven processing.
• Proven experience integrating third-party model APIs (such as OpenAI or Anthropic) and orchestrating complex external API state machines.
• Solid understanding of authentication, authorisation, and application security best practices, including token-based auth (JWT / OAuth 2.0).
• Hands-on experience with relational databases and SQL, and experience deploying services on at least one major cloud platform (AWS, GCP, or Azure) with containerization.
• Experience with the Model Context Protocol (MCP) or making enterprise systems agent-ready.
• Experience building RAG systems at scale: embeddings, vector databases, and grounded generation with citations.
• Familiarity with message brokers and streaming platforms: Kafka, RabbitMQ, or cloud-native equivalents.
• Experience with LLM evaluation, observability, and guardrail tooling.
• Experience with observability tooling (Prometheus, Grafana, OpenTelemetry, Datadog) and structured logging.

Responsibilities

• Agentic AI and LLM orchestration: design, build, and productionize agentic AI systems with multi-step, tool-using workflows and memory management.
• Implement retrieval-augmented generation (RAG) patterns with embeddings, vector stores, and grounded generation with citations.
• Apply prompt engineering, guardrails, evaluation, and fallback strategies to maximise reliability and minimise hallucination.
• Integrate model APIs (OpenAI/Anthropic) into backend services focusing on orchestration layers.
• Design multi-agent orchestration patterns for complex workflows.
• Backend Architecture and API Development: architect and develop robust, scalable backend services and RESTful/event-driven APIs.
• Lead end-to-end design of backend systems from domain modelling to service implementation.
• Build and maintain services with FastAPI and clean architecture; optimize performance and resource usage.
• Design async/concurrent processing, streaming interfaces (SSE/WebSockets) for real-time LLM updates.
• Implement reliable messaging (Kafka, RabbitMQ, Celery) and robust timeout, retry, and idempotent processing.
• Develop integration/middleware layers to decouple external dependencies and enable testability.
• Champion DevOps practices: CI/CD, observability, testing, and incident resolution.

Education

Bachelor's or Master's degree in Computer Science / Information Technology / Electronics or related engineering discipline

Tools

LangGraph

AutoGen

CrewAI

FastAPI

Kafka

RabbitMQ

Celery

Docker

Kubernetes

OpenAI API

Anthropic API

Job description

Responsibilities:

• Agentic AI and LLM Orchestration: Design, build, and take to production agentic AI systems with multi-step, tool-using workflows in which LLMs reason, plan, and trigger external function calls to complete complex tasks reliably. Architect and implement robust orchestration and tool-calling loops using frameworks such as LangGraph, AutoGen, or CrewAI, or well-structured custom implementations. Design conversational and workflow state management, memory, context handling, and multi-turn continuity for reliable long-running agent interactions.
• Implement retrieval-augmented generation (RAG) patterns: document ingestion, chunking, embeddings, vector stores, and grounded generation with citations to improve accuracy and traceability.
• Apply advanced prompt engineering, guardrails, evaluation, and fallback strategies to maximise reliability and minimise hallucination in production agent behaviour.
• Integrate model APIs (such as OpenAI and Anthropic) into backend services, focusing on the agentic routing and orchestration layer rather than training foundational models.
• Design agent-to-agent and multi-agent orchestration patterns for complex, decomposed workflows.
• Backend Architecture and API Development: Architect, design, and develop robust, scalable backend services and RESTful/event-driven APIs that power AI-driven and conventional workloads at scale. EngiNeo Solutions
• Lead the end-to-end design of backend systems from domain modelling and service.
• Build and maintain services using modern Python frameworks, with deep expertise in FastAPI, applying clean architecture and well-structured custom implementations.
• Own service performance tuning, profiling, query optimisation, caching, and resource management to ensure efficient handling of large-scale traffic.
• Design and implement asynchronous and concurrent processing patterns using async/await, background workers, task queues, and message-driven workflows.
• Build real-time streaming interfaces using Server-Sent Events (SSE) or WebSockets to stream LLM responses and live updates to clients.
• Implement reliable messaging and event-processing patterns using tools such as Kafka, RabbitMQ, or Celery where applicable.
• Define reliability standards, retry logic, timeout handling, graceful degradation, and idempotent processing across all asynchronous and AI workloads.
• Third-Party Integration and Orchestration: Design and build integrations with external APIs and platforms, managing complex multi-step orchestration flows and service-to-service state machines.
• Handle asynchronous third-party failure states gracefully alongside the agent's reasoning logic, with robust timeout, fallback, and reconciliation strategies.
• Build integration and middleware layers that abstract external dependencies and keep the core system decoupled and testable.
• Security, Authentication and Compliance: Design and implement secure authentication and authorisation, including token-based mechanisms such as JWT and OAuth 2.0 cryptographic signing, and role-based access control. Apply secure-by-design principles: input validation, secrets management, least-privilege access, and secure handling of sensitive data across services.
• Configure secure network patterns: CORS, content security policies, and outbound proxying and ensure services meet applicable security and compliance requirements.
• Cloud, DevOps and Reliability: Design and deploy backend and AI services on cloud platforms AWS, GCP, or Azure using containerization (Docker, Kubernetes) and cloud-native services.
• Champion DevOps practices: CI/CD pipelines, automated testing, observability, and environment management to improve reliability and deployment velocity. Implement monitoring, logging, tracing, and alerting to ensure system health, and lead incident diagnosis and resolution.
• Technical Leadership and Mentoring: Serve as a technical lead for the backend and AI engineering team, providing architectural guidance, conducting design and code reviews, and setting coding and documentation standards. Mentor and coach junior and mid-level engineers through hands-on guidance and structured feedback to accelerate their technical growth.
• Lead technical discussions, architecture design sessions, and proof-of-concept evaluations for new AI and backend initiatives. Contribute to engineering roadmap planning, sprint estimation, and technical backlog prioritisation in collaboration with the engineering manager.
• Collaboration and Stakeholder Engagement: Partner closely with product managers, frontend engineers, QA, data engineers, and solution architects to deliver high-quality, production-ready features. Engage directly with stakeholders to understand requirements, translate them into technical specifications, and manage expectations on delivery timelines. Participate in Agile ceremonies sprint planning, standups, retrospectives, and reviews and contribute to continuous improvement of the team's delivery processes.

Requirements:

• Bachelor's or Master's degree in Computer Science, Information Technology, Electronics, or a related engineering discipline.
• 7-10 years of hands-on backend engineering experience, with a strong focus on building and scaling production services and APIs.
• Proven, production-grade experience designing and building agentic AI / LLM-orchestration workflows using frameworks such as LangGraph, AutoGen, or CrewAI, or robust custom tool-calling implementations.
• Demonstrated ability to instruct an LLM to reliably reason, trigger external function calls, and manage conversational state and memory in production.
• Expert-level proficiency in Python, including asynchronous programming, with deep experience in FastAPI.
• Strong expertise in real-time streaming (SSE / WebSockets) for LLM responses and asynchronous/event-driven processing.
• Proven experience integrating third-party model APIs (such as OpenAI or Anthropic) and orchestrating complex external API state machines.
• Solid understanding of authentication, authorisation, and application security best practices, including token-based auth (JWT / OAuth 2.0).
• Hands-on experience with relational databases and SQL, and experience deploying services on at least one major cloud platform (AWS, GCP, or Azure) with containerization.
• Experience with the Model Context Protocol (MCP) or making enterprise systems agent-ready.
• Experience building RAG systems at scale: embeddings, vector databases, and grounded generation with citations.
• Familiarity with message brokers and streaming platforms: Kafka, RabbitMQ, or cloud-native equivalents.
• Experience with LLM evaluation, observability, and guardrail tooling.
• Experience with observability tooling (Prometheus, Grafana, OpenTelemetry, Datadog) and structured logging.

Soft Skills and Leadership Qualities:

• Strong ownership mindset takes end-to-end accountability for backend and AI deliverables and system reliability.
• Ability to communicate complex technical concepts clearly to both technical peers and non-technical stakeholders.
• Collaborative and team-oriented, works effectively across product, engineering, and business teams in a fast-paced environment.
• Structured problem-solver who can diagnose and resolve complex production issues under time pressure.
• Proactive learner who stays current with the rapidly evolving agentic AI, LLM, and backend engineering landscape.
• Experience working in Agile delivery environments with a continuous improvement mindset.