Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I am a software engineer with 4+ years of experience designing AI-enabled backend systems and distributed microservices across mobility, SaaS, and enterprise platforms. I currently develop production LLM applications at Uber using Python, FastAPI, OpenAI APIs, and RAG pipelines to automate driver support and policy decision workflows processing 3K+ daily tickets. I have built event-driven, multi-tenant backends using Java, Spring Boot, Kafka, PostgreSQL, and AWS for billing, authorization, and real-time data processing. My focus is on scalable API design, low-latency inference, audit-ready data handling, and production monitoring with Evidently AI and cloud observability tooling.…I am a software engineer with 4+ years of experience designing AI-enabled backend systems and distributed microservices across mobility, SaaS, and enterprise platforms. I currently develop production LLM applications at Uber using Python, FastAPI, OpenAI APIs, and RAG pipelines to automate driver support and policy decision workflows processing 3K+ daily tickets. I have built event-driven, multi-tenant backends using Java, Spring Boot, Kafka, PostgreSQL, and AWS for billing, authorization, and real-time data processing. My focus is on scalable API design, low-latency inference, audit-ready data handling, and production monitoring with Evidently AI and cloud observability tooling.

Vedant Mapare

Full Stack Developer, Back-End Developer, Web Developer, +5





I am a software engineer with 4+ years of experience designing AI-enabled backend systems and distributed microservices across mobility, SaaS, and enterprise platforms. I currently develop production LLM applications at Uber using Python, FastAPI, OpenAI APIs, and RAG pipelines to automate driver support and policy decision workflows processing 3K+ daily tickets. I have built event-driven, multi-tenant backends using Java, Spring Boot, Kafka, PostgreSQL, and AWS for billing, authorization, and real-time data processing. My focus is on scalable API design, low-latency inference, audit-ready data handling, and production monitoring with Evidently AI and cloud observability tooling.…I am a software engineer with 4+ years of experience designing AI-enabled backend systems and distributed microservices across mobility, SaaS, and enterprise platforms. I currently develop production LLM applications at Uber using Python, FastAPI, OpenAI APIs, and RAG pipelines to automate driver support and policy decision workflows processing 3K+ daily tickets. I have built event-driven, multi-tenant backends using Java, Spring Boot, Kafka, PostgreSQL, and AWS for billing, authorization, and real-time data processing. My focus is on scalable API design, low-latency inference, audit-ready data handling, and production monitoring with Evidently AI and cloud observability tooling.

Available to hire

I am a software engineer with 4+ years of experience designing AI-enabled backend systems and distributed microservices across mobility, SaaS, and enterprise platforms. I currently develop production LLM applications at Uber using Python, FastAPI, OpenAI APIs, and RAG pipelines to automate driver support and policy decision workflows processing 3K+ daily tickets.

I have built event-driven, multi-tenant backends using Java, Spring Boot, Kafka, PostgreSQL, and AWS for billing, authorization, and real-time data processing. My focus is on scalable API design, low-latency inference, audit-ready data handling, and production monitoring with Evidently AI and cloud observability tooling.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Work Experience

AI Backend Engineer (Python + LLM/RAG) at Uber

January 1, 2025 - Present

Architected a Python FastAPI RAG platform using OpenAI GPT-4 and pgvector, grounding responses for driver fare disputes, cancellations, and deactivation appeals across 150K+ policy documents, supporting 3K+ daily support tickets. Developed scalable LangChain embedding and ingestion pipelines processing 18M+ tokens/day, enabling policy and compliance updates to reach production search indexes within a 12-minute SLA. Implemented hybrid semantic retrieval, query rewriting, and reranking workflows using async Python, reducing hallucinated responses from 14% to 9% across 40K audited conversations. Integrated Evidently AI evaluation and monitoring pipelines analyzing 1M+ monthly LLM inference logs, detecting response drift during regional rule changes and preventing multiple production regressions. Automated refund eligibility and escalation recommendations using OpenAI function calling with schema-validated JSON outputs, reducing average case handling time from 9.5 to 6.8 minutes across 12

Software Engineer at Vivma Software Inc.

April 1, 2022 - July 1, 2023

Designed multi-tenant visit and authorization services in Java (Spring Boot) backed by Aurora PostgreSQL, adding sharded lookup tables that lifted read throughput 42% for high-volume agencies. Modeled scheduling and caregiver-management UI in React with selective prefetching and server-driven pagination; cut screen load time from 3 seconds to under 900 ms. Implemented real-time visit event streams using Kafka to sync EVV punches, shift edits, and plan-of-care changes; reduced stale-visit mismatches 31%. Created batch payroll and billing processors using Spring Batch + S3 to validate Medicaid file formats, provider rates, and rounding policies; held nightly processing inside 30 minutes across states. Set up AWS API Gateway + IAM-based access layers for external provider and aggregator integrations, tightening audit visibility and dropping integration-related errors.

Software Engineer at Sage Softtech

February 1, 2021 - March 1, 2022

Designed multi-tenant subscription and billing microservices using Java + Spring Boot with PostgreSQL, handling plans, usage rating, proration, and invoicing flows, reducing billing discrepancies by 23% across active customer accounts. Implemented event-driven background processing using Spring Boot async workers and Redis-backed queues to handle usage in gestion, notifications, and webhook retries, cutting peak-time processing delays by 80%. Optimized PostgreSQL performance through schema refactors, composite indexing, and query rewrites from Java repositories, improving p95 dashboard response time from 2.8s to 1.1s under concurrent load. Set up CI/CD pipelines for Spring Boot services using GitLab CI and Jenkins, enabling automated tests, versioned deployments, and safe rollbacks, increasing release frequency to 3–4 deployments per week with production regressions.