Kalahan Guntakala profile photo 1 on Tal

Kalahan Guntakala

AI/ML Engineer · LLM Pipelines & Agentic Workflows · Prompt Engineering · Full-Stack (MERN) · OpenAI & Claude APIs · RAG · n8n · BITS Pilani ’22

Hyderabad

4+ years of work experience

Nhancio
Oct 20241 yr 11 mosPresent

Remote

▸ Engineered LLM-powered agentic pipelines using CrewAI and MCP (Model Context Protocol) — integrating OpenAI, Anthropic, and AWS Bedrock APIs with multi-step orchestration logic, reducing client manual effort by ~70% and cutting end-to-end workflow execution time from hours to under 10 minutes. ▸ Built and deployed RAG pipelines using LangChain, ChromaDB, and OpenAI embeddings — implemented optimised chunking strategies and semantic caching, reducing LLM hallucination rate by ~60% and cutting query response time from ~8s to under 2s over client knowledge bases. ▸ Developed a vision-based browser automation agent (Playwright + LLM Vision) that interprets live screen-cast frames and autonomously executes browser actions from plain-English instructions — eliminated ~5 hours/week of repetitive manual browser tasks per client and enabled zero-code automation for non-technical users. ▸ Built an end-to-end OCR document intelligence pipeline using Python and OpenAI Vision API; applied structured prompt engineering achieving ~85% structured extraction accuracy on noisy documents; exposed via Flask REST API, processing 100+ documents per client engagement.
Stealth Startup
Dec 20239 mosAug 2024

▸ Architected and shipped a hyperlocal fashion e-commerce platform on the MERN stack — multi-vendor marketplace with vendor onboarding, product catalogue, order management, and try-before-you-buy fulfilment flow. ▸ Designed and integrated n8n + Zapier automation pipelines for vendor onboarding notifications, CRM triggers, and customer communication — cutting manual ops time by ~60% and enabling a team of 3 to operate at startup pace. ▸ Integrated OpenAI API for LLM-based product recommendations and automated customer query resolution; applied structured prompt engineering to reduce irrelevant responses and improve conversion. ▸ Owned complete product lifecycle — market research, system design, full-stack build, and GTM execution; built async REST APIs using Python/Flask for backend services; grew to initial vendor and customer base.
Storable
Mar 20239 mosNov 2023

Hyderabad

▸ Built a high-performance payment processing system integrating Payrix with Kafka event streaming — improved transaction success rates and reduced payment failures for self-storage operators serving thousands of US customers. ▸ Optimized full-stack architecture (React, TypeScript, Node.js, PostgreSQL, Docker) — reduced page load times by 30% through SQL query indexing, connection pooling, and lazy-loading strategies. ▸ Architected and delivered REST APIs for real-time, secure financial transaction workflows; enforced security compliance standards across all payment data flows. ▸ Owned on-call engineering rotation across production systems; reduced mean incident resolution time by 50% through structured root cause analysis, runbook documentation, and rapid hotfix deployment.
Kore.ai
Jun 20227 mosDec 2022

Hyderabad

▸ Integrated 3 enterprise third-party services (Confluence, Zendesk, SharePoint) into Kore.ai's AI-powered SearchAssist platform — enabling unified cross-source knowledge retrieval for large enterprise clients with thousands of daily active users. ▸ Designed secure REST APIs with encrypted data handling and role-based access control; wrote complex SQL queries for search analytics, improving pipeline efficiency and meeting enterprise data security requirements. ▸ Optimized Node.js + MongoDB backend via query refactoring, index tuning, and async response caching — measurably reduced API response times and improved throughput for AI-driven search queries. ▸ Collaborated with product and QA teams in Agile sprints to ship platform features on schedule, maintaining code quality through peer reviews and unit test coverage.
upGrad
Jan 20226 mosJun 2022

▸ Built a scalable, reusable UI component library using React.js and Next.js for one of India's largest ed-tech platforms, serving millions of active learners — components adopted across multiple product teams. ▸ Implemented JWT-based authentication and session management; integrated RESTful APIs and authored Jest unit tests, achieving high code reliability and enabling seamless continuous deployment.

Kalahan’s AI Stack

codex
codex
claudecode
claudecode

Education

BITS Pilani, Hyderabad Campus

Bachelor of Engineering - BE, Electronics and Instrumentation Engineering

BITS Pilani, Hyderabad Campus

20184 yrs2022

Postgraduate Degree, Leading with AI: Strategy to Business Impact

Hero Vired - IIMB & IBM

20251 yr2026

Join Kalahan and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Kalahan Guntakala on Tal

Headline

AI/ML Engineer · LLM Pipelines & Agentic Workflows · Prompt Engineering · Full-Stack (MERN) · OpenAI & Claude APIs · RAG · n8n · BITS Pilani ’22

About

I build things with LLMs , production-grade agentic pipelines, RAG systems, and AI-powered automation workflows that actually ship. With 2+ years of industry experience across AI engineering, backend systems, and full-stack development, I specialize in turning LLM capabilities into reliable, scalable products. My work spans: Agentic AI: Multi-step orchestration pipelines using OpenAI and Anthropic Claude APIs — automating research, extraction, and outreach workflows end-to-end. Vision Agents: Built a browser automation agent that interprets live screen-cast frames and executes actions from plain-English instructions (no hardcoded selectors, zero-code for non-technical users). Document AI: OCR + LLM pipelines for invoice and form extraction using OpenAI Vision API, with structured JSON output for downstream systems. Automation: n8n, Zapier, Make.com workflows connecting AI outputs to CRM, communication, and data systems cutting manual effort by 60–70%. I started my career at Kore.ai building enterprise search integrations, moved into SaaS at Storable (payment systems + Kafka), founded a MERN-based hyperlocal marketplace (TryNStyle), and now work as an AI Engineer at Nhancio AI delivering intelligent automation products. Currently deepening my ML foundations through a PG Program in Leading with AI (IIMB + IBM), with a focus on LangChain, LlamaIndex, and applied ML. Hyderabad | Open to AI/ML engineering roles, LLM product roles, and agentic AI projects

Location

Hyderabad

AI stack

codex

claudecode

Work experience

Artificial Intelligence Engineer at Nhancio, Oct 2024 - Present · 1 yr 11 mos, Remote, ▸ Engineered LLM-powered agentic pipelines using CrewAI and MCP (Model Context Protocol) — integrating OpenAI, Anthropic, and AWS Bedrock APIs with multi-step orchestration logic, reducing client manual effort by ~70% and cutting end-to-end workflow execution time from hours to under 10 minutes. ▸ Built and deployed RAG pipelines using LangChain, ChromaDB, and OpenAI embeddings — implemented optimised chunking strategies and semantic caching, reducing LLM hallucination rate by ~60% and cutting query response time from ~8s to under 2s over client knowledge bases. ▸ Developed a vision-based browser automation agent (Playwright + LLM Vision) that interprets live screen-cast frames and autonomously executes browser actions from plain-English instructions — eliminated ~5 hours/week of repetitive manual browser tasks per client and enabled zero-code automation for non-technical users. ▸ Built an end-to-end OCR document intelligence pipeline using Python and OpenAI Vision API; applied structured prompt engineering achieving ~85% structured extraction accuracy on noisy documents; exposed via Flask REST API, processing 100+ documents per client engagement.

Founder at Stealth Startup, Dec 2023 - Aug 2024 · 9 mos, ▸ Architected and shipped a hyperlocal fashion e-commerce platform on the MERN stack — multi-vendor marketplace with vendor onboarding, product catalogue, order management, and try-before-you-buy fulfilment flow. ▸ Designed and integrated n8n + Zapier automation pipelines for vendor onboarding notifications, CRM triggers, and customer communication — cutting manual ops time by ~60% and enabling a team of 3 to operate at startup pace. ▸ Integrated OpenAI API for LLM-based product recommendations and automated customer query resolution; applied structured prompt engineering to reduce irrelevant responses and improve conversion. ▸ Owned complete product lifecycle — market research, system design, full-stack build, and GTM execution; built async REST APIs using Python/Flask for backend services; grew to initial vendor and customer base.

Software Engineer at Storable, Mar 2023 - Nov 2023 · 9 mos, Hyderabad, ▸ Built a high-performance payment processing system integrating Payrix with Kafka event streaming — improved transaction success rates and reduced payment failures for self-storage operators serving thousands of US customers. ▸ Optimized full-stack architecture (React, TypeScript, Node.js, PostgreSQL, Docker) — reduced page load times by 30% through SQL query indexing, connection pooling, and lazy-loading strategies. ▸ Architected and delivered REST APIs for real-time, secure financial transaction workflows; enforced security compliance standards across all payment data flows. ▸ Owned on-call engineering rotation across production systems; reduced mean incident resolution time by 50% through structured root cause analysis, runbook documentation, and rapid hotfix deployment.

Associate Software Engineer at Kore.ai, Jun 2022 - Dec 2022 · 7 mos, Hyderabad, ▸ Integrated 3 enterprise third-party services (Confluence, Zendesk, SharePoint) into Kore.ai's AI-powered SearchAssist platform — enabling unified cross-source knowledge retrieval for large enterprise clients with thousands of daily active users. ▸ Designed secure REST APIs with encrypted data handling and role-based access control; wrote complex SQL queries for search analytics, improving pipeline efficiency and meeting enterprise data security requirements. ▸ Optimized Node.js + MongoDB backend via query refactoring, index tuning, and async response caching — measurably reduced API response times and improved throughput for AI-driven search queries. ▸ Collaborated with product and QA teams in Agile sprints to ship platform features on schedule, maintaining code quality through peer reviews and unit test coverage.

Javascript Developer at upGrad, Jan 2022 - Jun 2022 · 6 mos, ▸ Built a scalable, reusable UI component library using React.js and Next.js for one of India's largest ed-tech platforms, serving millions of active learners — components adopted across multiple product teams. ▸ Implemented JWT-based authentication and session management; integrated RESTful APIs and authored Jest unit tests, achieving high code reliability and enabling seamless continuous deployment.

Education

Bachelor of Engineering - BE, Electronics and Instrumentation Engineering, BITS Pilani, Hyderabad Campus

Postgraduate Degree, Leading with AI: Strategy to Business Impact, Hero Vired - IIMB & IBM