Headline

AI Engineer II @Swiggy | Prev. at DevRev

About

I am an AI Engineer II at Swiggy. I hold an Integrated M.Sc. degree in Artificial Intelligence and Machine Learning. I build agentic workflows, model orchestration frameworks, and machine learning architectures for high-traffic systems. At Swiggy, I design and deploy multi-agent systems and real-time inference pipelines. I optimize models to reduce system latency and cloud infrastructure costs. I manage project lifecycles from architecture design to production deployment. Previously at DevRev, I built custom AI agents and analytics tools. These tools processed operational data to give users actionable insights and automate workflows. Before DevRev, I worked at SuperAGI on open-source multi-agent systems. I led the development of SuperCoder 2.0. This tool achieved a 4th-place ranking on the SWE Lite Benchmark with 34% accuracy. I also designed agent communication protocols. These protocols decreased operational costs by 20% and improved response times by 65%.

Location

Bengaluru

AI stack

antigravity

codex

claudecode

Work experience

AI Engineer II at Swiggy, May 2026 - Present · 4 mos, Bengaluru, Karnataka, India, Architecting scalable LLM infrastructure and agentic systems to drive Swiggy's AI-native product vision.

Applied AI Engineer at DevRev, Nov 2024 - Apr 2026 · 1 yr 6 mos, Chennai, Tamil Nadu, India, - Collaborate with product, engineering, and customer success teams to identify pain points and translate business requirements into scalable AI solutions that deliver measurable impact across diverse customer segments - Deploy AI-powered automation solutions that integrate seamlessly with customer workflows, enhancing productivity through intelligent task optimization and process streamlining while maintaining system reliability and performance standards - Design and implement specialized AI agents built on foundation LLMs with robust validation frameworks, comprehensive guardrails, prompt engineering optimization, and real-time performance monitoring to ensure optimal functionality while maintaining focus on core objectives - Develop intelligent analytics visualizations that leverage advanced AI algorithms to transform complex data into actionable business insights, enabling data-driven decision making for enterprise customers and improving operational efficiency -Implement comprehensive LLM prompt engineering practices that refine input-output alignment, reduce hallucinations, improve factuality, and enhance response consistency across diverse customer interactions and edge cases - Lead technical implementation of customer-facing AI features, ensuring high-quality deliverables that meet enterprise standards for security, scalability, and user experience

AI Engineer at SuperAGI, Jan 2024 - Nov 2024 · 11 mos, Bengaluru, - Played a vital role in the development of Multi-Agent System architecture with Agent-to-Agent communication protocol. - Helped design the asynchronous subprocess communication among agents, which reduced the response time. - Taking complete ownership of projects from initial research and proof of concept through to final design and production deployment, collaborating closely with the CEO, CTO, and engineering teams. - Successfully developed SuperCoder2.0, which ranked fourth globally on the SWE Lite Benchmark, demonstrating exceptional technical prowess and innovation in AI tool development. - Boosted lead conversion rates through the implementation of automated outreach email generation, significantly impacting the sales pipeline.

Freelance at Freelance, Jan 2023 - May 2024 · 1 yr 5 mos, Remote, - Developed computer vision-based weapon detection system using object detection models for security screening applications, enhancing automated threat identification capabilities for safety-critical environments - Built intelligent job matching platform leveraging large language models and RAG architecture to automatically match optimal candidates to job descriptions, significantly improving recruitment efficiency and candidate-role alignment - Fine-tuned BERT model for cryptocurrency sentiment analysis, achieving production-ready performance with over 3 million lifetime downloads, demonstrating scalable NLP solution deployment for financial market analysis - Created time series forecasting model for Bitcoin price prediction using multi-asset data correlation, achieving high accuracy with $3-4 margin of error, enabling data-driven investment decision support - Delivered end-to-end AI solutions across security, recruitment, financial technology, and predictive analytics domains, consistently meeting client requirements while maintaining code quality and performance standards

Project Trainee at Bosch, Jul 2022 - Dec 2022 · 6 mos, Coimbatore, Tamil Nadu, India, Created Machine Learning tool for analyzing and predicting errors present in log files

Education

Master's degree, Artificial Intelligence and Machine Learning, Coimbatore Institute of Technology

Suguna RIP V School

Public links

github: https://github.com/KishoreKumar1308