Senior Software Engineer

Tekion; Bengaluru; On-site; INR 2,500,000 - 4,500,000; Full time

Benefits offered by this job

Tekion Career Growth

Job summary

Tekion is seeking an experienced software engineer to design and build production-grade AI platform services, including the AI gateway and MCP infrastructure. You will own end-to-end platform subsystems, optimize performance, and mentor junior colleagues in a fast-paced startup environment.

Required are 5+ years of Java-based experience, strong ownership, and a deep understanding of microservices, databases, and cloud technologies.

Qualifications

• Degree in CS or related field.
• 5+ years of software engineering experience.
• Strong ownership and code quality focus.
• Proficiency in Java and OO development.
• Experience building and scaling microservices.
• Knowledge of RDBMS and NoSQL technologies.
• Experience with enterprise backends and cloud-native apps.
• Familiarity with MCP design, LLM tooling, and AI workflows.

Responsibilities

• Design and build production-grade AI platform services and gateways.
• Lead the Tekion AI gateway design for multi-provider LLM access.
• Develop MCP infrastructure and agentic AI patterns.
• Own critical platform subsystems end-to-end.
• Resolve performance bottlenecks in inference and orchestration.
• Mentor junior and mid-level engineers.
• Collaborate with management, product, QA, UI/UX, and ML teams.
• Deliver high-quality work on a fast-paced startup cadence.

Skills

Java

Microservices

LLM fundamentals

Cloud platforms

Problem solving

Debugging & optimization

Code quality

Team mentorship

Education

Bachelor's/Master's in Computer Science

Tools

Spring Boot

Kafka

MongoDB

MySQL

AWS

ELK stack

Job description

Responsibilities:

• Design and build production‑grade AI platform services, LLM gateways, MCP servers, and agent orchestration frameworks.
• Lead the design and implementation of Tekion's AI gateway, a unified abstraction layer for multi‑provider LLM access with intelligent routing, fallback, cost optimization, token budgeting, and comprehensive observability.
• Design and implement MCP (Model Context Protocol) infrastructure, design agentic AI patterns, multi‑step reasoning workflows, tool orchestration, guardrails, and human‑in‑the‑loop escalation paths.
• Take end‑to‑end ownership of critical platform subsystems.
• Solve complex performance bottlenecks across inference and orchestration layers.
• Mentor junior and mid‑level engineers.
• Communicate and collaborate with management, product, QA, UI/UX, and AI/ML teams.
• Deliver with quality, on time, in a fast‑paced start‑up environment.

Requirements:

• Bachelor's/Master's in Computer Science or relevant fields.
• 5+ years of relevant software engineering experience.
• Strong sense of ownership.
• Excellent Java and object‑oriented development skills.
• Experience in building and scaling microservices.
• Strong problem‑solving skills, technical troubleshooting, and diagnosing.
• Expected to be a role model for young engineers with a strong sense of code quality and to enforce code quality within the team.
• Strong knowledge of RDBMS and NoSQL technologies.
• Experience in developing backends for enterprise systems (eCommerce / manufacturing/supply chain, etc.).
• Excellent understanding of debugging, performance, and optimization techniques.
• Experience with Java, MongoDB, MySQL, AWS technologies, ELK stack, Spring Boot, and Kafka.
• Experience in developing large‑scale cloud‑native applications.
• Strong understanding of LLM fundamentals.
• Experience with or strong knowledge of the MCP (Model Context Protocol) designing tool schemas, building MCP servers, tool discovery, and integrating MCP into agentic workflows.
• Understanding of AI gateway architecture, multi‑provider LLM routing, semantic caching, load balancing, token tracking, cost allocation, fallback chains, and request/response logging.
• Familiarity with agentic AI frameworks, such as LangChain/LangGraph, LlamaIndex, CrewAI, OpenAI Agents SDK, or similar orchestration tools.
• Experience with LLM observability and evaluation tracing (LangSmith, OpenTelemetry), LLM‑as‑judge evaluation, cost, and latency monitoring.
• Understanding of AI safety and guardrails input/output validation, PII detection, content filtering, and hallucination mitigation.
• Demonstrated ability to deliver in a fast‑paced environment.
• Good communication skills LLM‑powered systems: prompt injection, output validation, and PII handling.