Abhay Tuduma
Open to work3 years experienceGenAI Developer
GenAI Developer | LLMs | RAG Pipelines | LangChain | Agentic AI | Backend Engineering
Standing Rank
Rank Not Available
Developer Badges
No badges earned yet
Skills & Technologies
53 skillsGitHub Activity Evidence
GitHub Footprint Not Connected
Recruiters value verified code contributions. Linking a GitHub account displays live heatmaps, active contribution metrics, repository languages, and push activity tags.
Work Experience
AI Engineer
SpurTree Technologies
- Architected a production-grade conversational RAG pipeline (OCR → chunking → embedding → Azure Cognitive Search → GPT-4) integrated with a multi-turn chatbot interface; processed 50,000+ enterprise documents, cutting per-batch processing time from ~45 min to under 18 min — a 60%+ throughput gain.
- Built LangChain-powered agent orchestration with LangGraph for stateful multi-turn conversational flows and cross-document reasoning; chatbot handled dynamic follow-up questions, clarifications, and context- aware responses across long sessions.
- Implemented Function Calling / tool-augmented agents for summarization, cross-document comparison, and translation within the conversational AI assistant; reduced manual document review by 50% for enterprise teams.
- Engineered optimised chunking strategy and vector embedding pipeline with Pinecone and Azure AI Search; improved retrieval relevance by 40%+ validated across 200+ test queries.
- Implemented MCP-style context injection, prompt versioning, and AI output schema validation on Azure AI Foundry; used few-shot prompting and structured output enforcement to reduce LLM hallucination by 40%+ in production, measured via format compliance logs.
- Integrated Langwatch for LLM conversation monitoring — tracked response quality, token usage, and latency; drove iterative prompt and retrieval improvements using evaluation metrics.
- Developed middleware and backend logic using FastAPI to support dynamic, multi -turn conversational flows with context window management and memory strategies (conversation buffer + summary memory).
- Led end-to-end backend ownership of a 4-domain SaaS platform across a team of 3–5 engineers; shipped 15+ features and compressed time-to-market by 30% through sprint planning and stakeholder alignment.
- Designed modular, extensible backend architecture enabling AI-driven workflows across 4 domains — new modules onboarded without re-architecting core services, reducing future engineering overhead.
- Built scalable REST APIs with Webhooks integration and OAuth-based authentication for secure third-party service connectivity across multiple consumer-facing applications.
- Mentored 4 intern engineers — conducted weekly code reviews, pair programming sessions, and guided them through LLM integration, RAG concepts, and backend API design; accelerated their onboarding by 40%.
- Delivered LLM-powered features including semantic search, NER-based content tagging, and AI-driven recommendation flows across EdTech and Real Estate modules.
- Engineered a granular role-user-permission schema with JWT token optimisation across distributed microservices; reduced authentication failures by 80%, verified via production error logs, serving 50+ enterprise users.
- Implemented OAuth flows and Webhook event listeners for third-party integrations; delivered a React admin dashboard eliminating permission-change bottlenecks entirely.
- Built an event-driven microservices backend with LLM-integrated content pipelines and function-calling agents for automated product tagging and description generation; sustained zero downtime during peak traffic events.
- Accelerated deployment cycles with CI/CD pipelines (Docker + Azure DevOps); owned monitoring and incident response for all backend systems using token usage and latency dashboards.
Projects
GEMS AI — Conversational Document Intelligence Chatbot
- Multi-turn conversational AI assistant over 50,000+ enterprise documents; LangChain + LangGraph agent orchestration with 60%+ batch processing time reduction and 40% LLM hallucination reduction in production.
Enterprise RBAC & Admin Platform
- Granular role-user-permission schema with JWT + OAuth auth across multi-repo microservices; 80% reduction in auth failures measured via production logs.
Live Commerce Platform — Scalable AI Backend
- Event-driven microservices with LLM-integrated content pipelines and function-calling agents for automated workflows; zero downtime during high-velocity live commerce events.
Leaderboard Standings
Leaderboard Position Pending
Global test scores, peer standing percentiles, and algorithm leaderboard ranks are updated dynamically.
Assessment Highlights
Assessments Not Completed
Coding evaluations, system assessment results, and conceptual score badges will appear here after taking a test.
AI Collaboration Score
AI Collaboration Score Pending
Developer coding behavior, assistant cooperation, and AI pair-programming indicators are evaluated during live coding sessions.
Role Compatibility Profile
Role Compatibility Analysis Pending
Custom matching reports, candidate role compatibility percentiles, and core engineer strength profiles are processed once conceptual code screenings are complete.
Achievements
Batch Processing Optimization
60% reduction in batch processing time on a 50,000+ document AI chatbot platform.
Authentication System Improvement
80% reduction in authentication failures.
LLM Hallucination Reduction
40% reduction in LLM hallucination rates.
About Details
Professional Bio
GenAI Developer with 2.8 years of production experience building LLM-integrated systems, RAG pipelines, and multi-agent agentic workflows. Proven track record of delivering measurable impact, including significant reductions in processing time, authentication failures, and LLM hallucination rates.
B.E. Computer Science in Computer Science
JSS Academy of Technical Education, Bengaluru (2019 - 2023)