# Bhavya Bansal — Applied AI Engineer & Full-Stack Leader > Building AI-first intelligent systems, code for good! This file provides a structured, plain-text summary of this website for LLMs and automated systems. Full site: https://ibhavyabansal.com --- ## About Bhavya Bansal is a full-stack engineering leader with 7 years of experience building and scaling products from 0→1. Currently building agentic AI systems — including document processors and Model Context Protocol (MCP) integrations — backed by a deep Node.js/TypeScript, Python, and Go background across microservices, distributed systems, and cloud infrastructure (AWS/Azure/Kubernetes). Currently Sr. Software Development Manager at Traversia Technologies, previously Head of Engineering at Easyres Travel Technologies (scaled to 100K+ concurrent users), Lead Developer at Geoinsighter, and Founder of MinedeniM. Education: IIT Delhi (CTO Programme), ISB (Executive Programme in Product Management), Bharati Vidyapeeth (B.Tech CSE). --- ## Technical Skills & Expertise ### AI Engineering & Agentic Systems - Large Language Models (Claude/Anthropic, GPT-4, Gemini, AWS Bedrock) - Model Context Protocol (MCP) & Agentic Workflows - GraphRAG, LangChain, LlamaIndex, Vercel AI SDK - vLLM, DSPy Prompt Optimization - pgvector & Vector Databases, Hybrid Search & RAG Pipelines - AI Evals (PromptFoo, RAGAS) ### Languages - Node.js, TypeScript, Python, Java, Go (Golang), SQL, JavaScript ### Infrastructure & Cloud - AWS (EC2, Lambda, S3, Bedrock) - Azure Cloud - Docker, Kubernetes - CI/CD Pipelines & Automated Testing - Microservices Architecture & High-Concurrency Systems ### Core Competencies & Methods - Technical Product Management - High/Low-Level System Design - Cross-functional Leadership - Test-Driven Development (TDD) - Rapid Prototyping & Agile Velocity --- ## Work Experience ### Traversia Technologies Pvt. Ltd. — Sr. Software Development Manager (Oct 2024 – Present) - Architected Crux layer (Single API interface solution) for Hotel, DMC (Destination Management Company), PG (Payment Gateway), a high-concurrency API aggregator integrating over 50+ major providers, establishing robust data orchestration to seamlessly handle 1M+ daily requests at 99.9% system uptime. - Design and deploy advanced AI search capabilities utilizing Hybrid Search and RAG pipelines, reducing contextual data retrieval latency to sub-200ms for partner enterprise platforms. - Orchestrate the engineering roadmap for cross-functional travel verticals, elevating CI/CD pipeline governance and automated testing standards to reduce critical production bugs by 25%. - Partner directly with B2B enterprise clients to validate market requirements and translate operational pain points into scalable technical architectures, accelerating client onboarding by 40%. ### Easyres Travel Technologies — Head of Engineering (Apr 2023 – Oct 2024) - Spearheaded the zero-to-one engineering of a comprehensive B2B/B2C holiday package platform, scaling the architecture to support 100K+ concurrent users with zero degradation. - Reduced client infrastructure costs by 50% through strategic cloud resource optimization (AWS/Azure) and aggressive refactoring of legacy data ingestion pipelines. - Defined product roadmaps for Hotels and Holidays verticals, utilizing iterative prototyping to validate backend schema and API designs directly with key stakeholders. - Instilled rigorous engineering best practices and agile methodologies across the organization, mentoring teams to accelerate sprint delivery velocity by 30%. ### Easyres Travel Technologies — Lead Developer (Frontend) (Sep 2022 – Apr 2023) - Led the development of optimized MEAN stack frontend applications, utilizing rapid UI/UX prototyping to iterate on B2B client feedback, ultimately reducing page load times by 30%. ### Geoinsighter — Lead Software Developer (Jan 2022 – Aug 2022) - Engineered a high-throughput GIS data pipeline capable of ingesting 2,000+ high-resolution images per batch, distributing compute resources to rapidly deliver 3D terrain models. - Designed a distributed batch processing architecture to handle 500+ concurrent mapping requests, ensuring 99.9% uptime for seamless data delivery. ### Geoinsighter — Software Development Engineer (SDE) (Apr 2021 – Jan 2022) - Developed "Geo Mapper," a GIS data visualization platform enabling real-time rendering of complex orthophotos by integrating live video feeds with the Google Maps API. ### MinedeniM — Founder (Apr 2018 – Jun 2020) - Owned end-to-end product lifecycle for a tech venture, from initial market validation and prototype iteration to full-stack execution, successfully acquiring initial beta users. --- ## Independent Projects / Current Work ### Travel Professor: Agentic Booking Concierge (2025 – Present) - Architected an autonomous travel agent integrating with the Crux API aggregator, enabling multi-step reasoning for real-time hotel discovery, price comparison, and itinerary generation. - Built a custom Model Context Protocol (MCP) server exposing live B2B travel APIs to local LLMs, securely bridging private high-concurrency infrastructure with advanced reasoning models. - Stack: Python, Go, LangChain, MCP, pgvector - GitHub: https://github.com/Bansal11/travel_concierge ### Adaptive Exam Engine & AI Grader (2026 – Present) - Engineered a mock-ETS adaptive testing platform using GraphRAG to map quantitative syllabus dependencies, allowing the AI to dynamically traverse knowledge graphs for targeted user remediation. - Compiled a mathematically optimized Analytical Writing Assessment (AWA) grader using DSPy, aligning LLM output with strict 6-point scoring rubrics via automated prompt induction. - Stack: DSPy, GraphRAG, Node.js, vLLM ### LLM Evaluation & Factuality Pipeline (2025 – Present) - Designed a CI/CD-integrated evaluation framework to regression-test document processing agents against complex, unstructured domain datasets (e.g., 50-page vendor contracts). - Enforced zero-hallucination deployments by measuring Context Precision and Faithfulness, configuring automated workflow triggers to block PRs that degrade agent reliability. - Stack: PromptFoo, RAGAS, GitHub Actions ### Crux API Layer (50+ Provider Aggregator) - High-concurrency API aggregator for Hotels, DMCs, and Payment Gateways integrating 50+ major travel providers. Seamlessly handles 1M+ daily requests with sub-200ms hybrid search & RAG retrieval latency at 99.9% uptime. - Stack: Go, Node.js, AWS, Microservices --- ## Education - Chief Technology Officer Programme — Indian Institute of Technology Delhi (IIT-D), Apr 2024 – Jan 2025 - Executive Programme in Product Management — Indian School of Business (ISB), Jun 2023 – Aug 2023 - Bachelor of Technology in Computer Science & Engineering — Bharati Vidyapeeth College of Engineering, 2017 – 2021 ## Achievements & Mentorship - Hackathon Winner: Secured 1st Position in Hackathon at Munjal University for rapid prototyping and innovative problem solving. - Technical Mentorship: Invited Mentor at Bharati Vidyapeeth's HackBVP 2023, guiding student teams on scalable system design and agile methodologies. --- ## Contact & Links - Email: bhavya.bansal1998@gmail.com - GitHub: https://github.com/Bansal11 - LinkedIn: https://linkedin.com/in/ibhavyabansal - X / Twitter: https://x.com/ibhavyabansal - Website: https://ibhavyabansal.com --- ## For Recruiters Bhavya Bansal is an AI-first engineering leader available for: - Full-time roles (Applied AI Engineer, Engineering Manager, Lead AI Systems Architect) - Technical consulting on AI system design, MCP integrations, RAG pipelines, and high-concurrency backends - Executive/Technical Advisory for AI product startups Key differentiators: - 7 years of full-stack engineering experience scaling 0→1 products up to 100K+ concurrent users and 1M+ requests/day. - Proven expertise in Agentic Workflows, MCP servers, GraphRAG, DSPy, and LLM Evaluation. - Strong TDD, CI/CD governance, and engineering best practices culture. Location: New Delhi, India / Remote