Ajeeth Kumar

AI Engineer with 5+ years of overall professional experience, including 2+ years of hands-on experience designing and deploying production-grade Generative AI and Agentic AI solutions. Experienced in building multi-agent systems, RAG pipelines, intelligent document processing, LLM applications, and AI-powered automation using Python, FastAPI, LangGraph, AWS Bedrock, LangChain, Redis, Kafka, PostgreSQL, and Docker. Strong backend engineering foundation with expertise in LLM orchestration, prompt engineering, embeddings, vector search, asynchronous processing, microservices, and cloud-native AI platforms.

Experienced in taking AI solutions from architecture and development through evaluation, deployment, observability, and production optimization on AWS.

AI Engineer - Annular Technologies

(2024-05)

GTMer – Enterprise AI Sales Development Platform

• Designed and developed a production-grade, multi-tenant Enterprise AI SaaS platform automating B2B sales workflows including AI-driven lead enrichment, email generation, voice calling, chatbot automation, CRM workflows, competitor intelligence, and analytics.
• Architected Agentic AI workflows using LangGraph and AWS Bedrock, implementing multi-agent orchestration, state management, tool calling, workflow routing, structured outputs, retries, and failure recovery.
• Built enterprise RAG and document intelligence pipelines for unstructured data using embeddings, vector search, OCR, document processing, prompt engineering, and LLM-based extraction with models such as Qwen and Mistral.
• Developed scalable Python/FastAPI microservices with PostgreSQL, Redis, Kafka, and Celery, supporting asynchronous processing, distributed workflows, concurrency, and reliable API operations.
• Implemented LLMOps and observability using LangSmith for workflow tracing, LLM evaluation, token/cost monitoring, debugging, and structured-response validation.
• Containerized and deployed AI services on AWS using Docker, with production infrastructure involving EC2, S3, CloudFront, and Nginx, incorporating rate limiting, timeouts, retries, and idempotent processing.
• Collaborated with Product, QA, and enterprise stakeholders throughout the AI product lifecycle, covering architecture, development, testing, model evaluation, deployment, monitoring, and continuous improvement.

Autonomous AI Agent Platform for Code Intelligence & Self-Healing Systems - Annular Technologies

(2024-05)

• Built an autonomous AI/ML platform using Python and FastAPI, applying NLP, LLMs, and Transformer-based models for code understanding, intelligent reasoning, automation generation, and decision-making.
• Engineered a Graph-RAG and semantic reasoning pipeline combining vector embeddings, knowledge graphs, semantic similarity, and multi-factor scoring to support context-aware AI/ML reasoning across large codebases.
• Integrated and orchestrated Generative AI and Deep Learning-based foundation models through AWS Bedrock, LiteLLM, and LangChain, using structured prompting, tool calling, validation, and fallback strategies for reliable agent execution.
• Developed AST-based code intelligence using ts-morph and Tree-Sitter, combining deterministic program analysis with LLM-based semantic understanding to analyze TypeScript/React applications and validate generated automation artifacts.
• Designed a self-healing AI agent that consumes execution telemetry, performs automated failure diagnosis using multi-source reasoning, and selects remediation strategies, reducing manual maintenance effort by 80%+.
• Architected a domain-driven microservices platform across Analysis, Reasoning, Execution, and Knowledge domains, integrating PostgreSQL, Playwright, Jira, and Kiwi TCMS for end-to-end intelligent automation.
• Implemented an end-to-end AI/ML and Agentic AI lifecycle covering data/context ingestion, embedding generation, retrieval, model inference, evaluation, validation, automation execution, and continuous failure recovery.

Technical Recruiter - Reveille Technologies

(2022-10 - 2023-07)

• Managed end-to-end technical recruitment lifecycle — sourcing, screening, and interviewing candidates for engineering, backend, and enterprise technology roles.
• Partnered with hiring managers and engineering teams to define role requirements, build candidate pipelines, and conduct technical screening rounds for Python, SQL, and backend engineering roles.
• Coordinated offer negotiations, onboarding logistics, and documentation, ensuring smooth candidate-to-hire transitions in collaboration with product and engineering stakeholders.

Customer Support - SpiceJet Ltd

(2019-11 - 2021-12)

• Provided production support for enterprise reservation and booking systems.
• Investigated application, database, and infrastructure issues while ensuring high system availability.
• Performed root cause analysis and collaborated with engineering teams to resolve production incidents.

Master of Computer Applications - AI & ML - Chandigarh University (2024 - 2026)

Bachelor of Science - Computer Technology - Bharathiar University (2015 - 2018)