Tharika Vikas Srinivasa — Software Engineer

I’m a Software Engineer at A.P. Moller Maersk working at the intersection of backend systems and applied AI, with a focus on building production-grade LLM-powered systems. Over the past couple of years, I’ve worked on designing and scaling intelligent systems that go beyond simple integrations—solving real-world problems involving ambiguity, scale, and reliability. My recent work includes: • Building an AI-powered support automation system that processes thousands of customer emails across global regions. The system uses SOP-driven agent workflows, dynamically selects actions based on country and case type, and orchestrates 30+ internal tools to generate high-quality response suggestions for human agents. • Solving key challenges in production AI systems such as: – reducing context bloat in LLM workflows – eliminating looping and non-deterministic agent behavior – handling large, non-optimized API responses – improving reliability via composite tools and structured execution • Designing and developing a platform-wide chatbot covering 15+ products, using advanced semantic indexing techniques for retrieval instead of standard vector-based approaches. Alongside AI systems, I’ve also worked on: • Backend development using Spring Boot (Java) • API design and integration across multiple internal platforms • CI/CD pipelines, Docker-based deployments, and Kubernetes environments • Observability using Grafana and custom health monitoring systems • GitOps workflows and automation using Python I’m currently focused on deepening my understanding of distributed systems, infrastructure, and scalable AI architectures. I’m particularly interested in roles in Applied AI, AI Platform Engineering, and backend systems where I can build reliable, production-ready intelligent systems at scale.

Stackforce AI infers this person is a Backend Engineer specializing in Applied AI and intelligent systems for SaaS.

Location: Bengaluru, Karnataka, India

Experience: 1 yr 11 mos

Skills

Backend Development
Applied Ai
Conversational Ai

Career Highlights

Expert in building LLM-powered intelligent systems.
Proven track record in backend development and API integration.
Strong focus on solving production AI challenges.

Work Experience

A.P. Moller - Maersk

Associate Software Engineer (1 yr 11 mos)

Intern (11 mos)

Education

Bachelor of Engineering - BE at RNS Institute of Technology - India

at ALLEN

at National Hill View Public School

Tharika Vikas Srinivasa

Software Engineer

Bengaluru, Karnataka, India1 yr 11 mos experience

AI EnabledAI ML Practitioner

Key Highlights

Expert in building LLM-powered intelligent systems.
Proven track record in backend development and API integration.
Strong focus on solving production AI challenges.

Stackforce AI infers this person is a Backend Engineer specializing in Applied AI and intelligent systems for SaaS.

Contact

Skills

Core Skills

Backend DevelopmentApplied AiConversational Ai

Other Skills

Spring BootApache CamelAI-powered support automationSOP-driven workflowsAPI integrationDockerKubernetesBigQuerySemantic indexingChatbot developmentBackend APIsRasaLangChainGitHub ActionsGrafana

About

Experience

1 yr 11 mos

Total Experience

1 yr 11 mos

Average Tenure

1 yr 11 mos

Current Experience

A.p. moller - maersk

2 roles

Associate Software Engineer

Jul 2024 – Present · 1 yr 11 mos · Bengaluru, Karnataka, India · Hybrid

Designed and built an AI-powered support automation system to assist human agents in responding to thousands of global customer emails
Implemented SOP-driven agent workflows that dynamically select execution paths based on country and case type (RFC), enabling structured and reliable automation
Integrated and orchestrated 30+ internal tools (APIs across multiple products) within the agent framework
Solved critical production challenges in LLM systems including:
1. reducing context bloat from large API responses
2. eliminating looping and non-deterministic agent behavior
3. handling invalid argument and tool execution errors
4. improving determinism via composite tools and structured workflows
Built a platform-wide intelligent chatbot covering 15+ products using advanced semantic indexing techniques for document retrieval (beyond standard vector database approaches)
Designed backend APIs and evaluation pipelines to monitor and improve chatbot performance
Developed and enhanced RESTful APIs using Spring Boot (Java) for scalable integrations
Built and configured Apache Camel routes for workflow orchestration
Gained hands-on exposure to cloud and infrastructure:
1. worked with BigQuery and GCP storage for data-driven decision making
2. deployed services using Docker and Kubernetes
3. managed secrets using Vault and external secrets

Spring BootApache CamelAI-powered support automationSOP-driven workflowsAPI integrationDocker+4

Intern

Jul 2023 – Jun 2024 · 11 mos · Bengaluru, Karnataka, India · Hybrid

Built conversational AI prototypes using Rasa and LangChain, exploring architectures for enterprise automation
Contributed to CI/CD pipeline setup using GitHub Actions
Developed and enhanced backend APIs using Spring Boot
Built observability systems using Grafana and Prometheus for real-time monitoring
Designed and implemented a broker health monitoring system using Python:
1. validated message flow across multiple ActiveMQ Artemis brokers
2. exported metrics via Telegraf and visualized system health in Grafana
3. Improved system reliability through comprehensive JUnit testing