VINAYAKA HEGDE
Explore with AIResume

Vinayaka Hegde

AI Engineer (multi-agentic, tool calling, FastAPI, AWS Certified)

College Station, TX

I architect autonomous, agentic systems, which is a serious way of saying: I build things that think and act for you, so you can focus on the stuff that actually needs a human

Designing and shipping agentic AI in production: multi-agent orchestration, model routing, autonomous workflows, and deploying to AWS

Vinayaka Hegde

Featured Builds

Things I've built end-to-end where AI, infra, and product thinking meet.

Signalsly

Signalsly

College Station, TX
Mar 2026 - Jun 2026

Agentic AI incident triage assistant that helps engineers resolve production issues faster by halving mean-time-to-resolution.

  • Built Signalsly, an agentic incident triage system using LangGraph, OpenSearch, and AWS Bedrock, reducing PagerDuty mean-time-to-resolution (MTTR) by 50%.
  • Engineered a hybrid RAG pipeline using Titan Embed v2, OpenSearch BM25, RRF, and Amazon Rerank v1, improving top-k retrieval precision by 30%.
  • Developed asynchronous FastAPI APIs for real-time incident-triage, reducing API latency by 15%.
  • Implemented AWS Bedrock Guardrails and contextual grounding checks to mitigate prompt-injection attacks.
LangGraphAWS BedrockOpenSearchFastAPI
RAG Portfolio

RAG Portfolio

College Station, TX
May 2026 - Jun 2026

Built a low-latency RAG backend powering a context-aware portfolio chatbot over personal data.

  • Built a production-grade RAG system using FastAPI, ChromaDB, SQLite, and Groq Llama 3.3 70B, enabling sub-350ms end-to-end response latency for grounded portfolio chat.
  • Designed a semantic retrieval pipeline with all-MiniLM-L6-v2 embeddings and vector search, achieving 2–5ms query-time retrieval over ~50MB index.
  • Implemented SSE-based streaming for LLM responses, enabling ~200–300ms first-token delivery and improved perceived responsiveness.
  • Architected a RAG backend with automated ingestion, chunking, and indexing across resume, projects, and mindset data for multi-source contextual retrieval.
FastAPIChromaDBGroq Llama 3.3SSEDockerRender
miniHive

miniHive

College Station, TX
Mar 2026 - Jun 2026

Distributed SQL query engine on Hadoop MapReduce supporting SELECT, WHERE, and aggregations on HDFS datasets.

  • Engineered miniHive, a distributed SQL engine on Hadoop MapReduce executing SELECT, WHERE, and aggregations via custom Python mapper-reducer pipelines.
  • Designed an HDFS-backed schema registry using index files for persistent table metadata, enabling dynamic column resolution and type-aware execution.
  • Implemented predicate pushdown at the map phase to filter rows before shuffle, minimizing network I/O.
  • Built a SQL parser from scratch for DDL/DML, dynamically constructing Hadoop Streaming jobs with resolved column indices and inferred datatypes.
PythonHadoopMapReduceHDFSSQL

Experience

Roles where product thinking, customer obsession, and AI delivery all had to work together.

Verce Solutions

Verce Solutions

Bangalore, India
Jan 2023 - Jul 2024

SDE 2

WHAT I DID
  • Architected a cloud-based DBaaS backend for a ride-sharing platform, integrating user and ride microservices via REST APIs
  • Containerized services with Docker and built a Flask orchestrator managing primary/replica worker containers with dynamic read/write separation
  • Implemented RabbitMQ RPC communication between orchestrator and workers, enabling fully async read/write handling at scale
  • Built request-volume based auto-scaling of primary/replica workers via Docker SDK and threading, ensuring high availability under peak load
WHAT THE IMPACT WAS
  • Delivered a fault-tolerant distributed DBaaS backend for a ride-sharing platform, directly unblocking pilot launch
  • Achieved 99%+ uptime and sub-100ms latency through primary/replica auto-scaling and async RabbitMQ-based request handling
  • Eliminated single points of failure via read/write worker separation, reducing DB bottlenecks by 80% under peak concurrent load
PythonFlaskDockerRabbitMQREST APIs
LeadSquared

LeadSquared

Bangalore, India
Mar 2022 - Dec 2022

Software Engineer

WHAT I DID
  • Owned the entire auth lifecycle (8+ flows) for LeadSquared's Marketplace and App Connector Platform using Java, Spring Boot and Spring Security
  • Implemented refresh token rotation with hashed storage and server-side validation, eliminating token reuse and replay attacks
  • Built 15+ RESTful endpoints across Auth, User and Connector modules, managing schema via Flyway across 10+ PostgreSQL tables
  • Enforced multi-tenant access control via Spring Security's SecurityContext, ensuring zero cross-tenant data leakage
WHAT THE IMPACT WAS
  • Delivered production-grade auth infrastructure for a Series C SaaS (500+ employees)
  • Reduced auth vulnerabilities by 90% through token rotation, hashed storage and server-side validation
  • Enabled 50+ customers to securely connect third-party apps with zero auth failures
JavaSpring BootSpring SecurityPostgreSQLFlyway
Hewlett Packard Enterprise

Hewlett Packard Enterprise

Bangalore, India
Jan 2020 - Jun 2020

Project Intern

WHAT I DID
  • Built an internal QA automation tool for HPE Cloud Services using Python and AWS (Lambda, S3, EC2, IAM) to run automated tests across cloud endpoints
  • Designed test pipelines executing 200+ automated test cases across 15+ internal cloud services, replacing manual QA validation
  • Configured AWS IAM roles and policies to enforce secure access control across the QA tool's infrastructure
  • Collaborated with 8 engineers in agile sprints, contributing to architecture decisions, code reviews, and demos
WHAT THE IMPACT WAS
  • Reduced QA cycle time by 70% by replacing manual testing with a fully automated cloud-native platform
  • Improved release confidence across 15+ HPE Cloud services by catching critical defects pre-deployment
  • Delivered a production-ready tool adopted by the Cloud Services team within the 6-month internship
PythonAWS LambdaS3EC2IAMAgile

Skills

Areas where I can contribute quickly and own the work end to end.

Languages & Backend

PythonJavaFastAPIFlaskPydanticSQLAlchemyREST APIsJWT AuthAsync APIs

AI Stack

LangGraphRAGLLMsLLM OrchestrationMulti-step Tool CallingAWS Bedrock (Claude, Titan, Rerank)Titan Embed v2Amazon Rerank v1Bedrock GuardrailsOpenSearch (KNN, BM25)Hybrid Retrieval (RRF)Vector SearchEmbeddings

Databases

SQLPostgreSQLMongoDBDynamoDB

Cloud & DevOps

AWS (EC2, S3, IAM, CloudWatch)DockerRabbitMQGitGitHub ActionsCI/CDPostman

Certifications

AWS Certified Developer – Associate, 2025