Siva Datta Koneru
Verified Expert in Engineering
AI and Back End-heavy Full-stack Developer
Hyderabad, India
Toptal member since August 17, 2026
Siva architected Panviva's enterprise RAG platform for 11,000+ users, cutting AI infrastructure costs 35% and average response time 20%. A senior engineer with 9+ years building production AI platforms, back end, and distributed systems, he also built Bloomberg fixed-income analytics handling 100M+ market/trade records daily and production AI/MCP-based automation at GoodNotes. His expertise spans RAG, LLM evaluation, AI agents, Python/FastAPI, C#/.NET, React, and Azure.
Portfolio
Experience
- .NET - 9 years
- REST APIs - 9 years
- SQL - 8 years
- Python - 8 years
- React - 5 years
- Azure - 4 years
- Claude - 2 years
- Retrieval-augmented Generation (RAG) - 2 years
Work Experience
Full-stack Software Engineer
GoodNotes
- Implemented production AI image generation and editing, reaching approximately 5% of Text Documents users during its first month of deployment.
- Built an image Q&A using a vision LLM, with approximately 12% of Text Documents users using the capability, and responses taking approximately 3 seconds.
- Resolved 100+ bugs across web, back end, and iOS, reducing crashes by 80% and clearing all high-priority product-tagged bugs.
- Reduced manual triage and task-preparation effort by 40% across approximately 4–5 Jira tickets per day; the automated workflow reduced the relevant work from roughly half a day per ticket to about 1 hour per ticket.
- Developed Python and FastAPI back-end services and TypeScript and React front-end functionality for AI features and product improvements.
- Partnered with engineering, product, and design teams to investigate issues, scope fixes, and deliver production-ready releases.
Senior Software Engineer
Upland Software
- Served as the primary engineer and architect for major components of Panviva's enterprise knowledge platform, supporting 11,000+ registered users.
- Reduced average AI response time by 20%, achieving a 99.9% prompt-cache hit rate and 3.5-second p95 AI-request latency.
- Built LLM-as-judge evaluation across 5 tenants using an authored golden set of approximately 80–100 Q&A/citation sets to evaluate RAG answer quality.
- Designed multiple production AI workflows, including a prompt-enhancement → answer-generation → hallucination-check pipeline, plus separate summarization and title-generation workflows.
- Reduced AI infrastructure costs by 35% through prompt caching and token optimization.
- Deployed self-hosted Phi-3 inference on an Azure NC-series A100 GPU VM for three tenants at approximately $80/hour continuously, then consolidated on Azure OpenAI at approximately $120/day on cost grounds.
- Modernized a singleton .NET Framework/ASP.NET Blazor application into modular .NET and Python/FastAPI microservices with React, leading a 4-engineer squad; two engineers reached mid-level roles within 18 months.
- Reduced microservice update cycles from biweekly full releases to independent service updates and reduced error-resolution time from approximately 1 month to 1 week.
- Implemented multi-tenant architecture, RBAC, audit logging, AWS Bedrock Guardrails, and regex redaction of phone and credit-card data for 20+ enterprise customers across the US and APAC, including GDPR- and HIPAA-sensitive environments.
- Defined architecture standards across a 12-person engineering organization, onboarded 5 developers, and configured RAG observability that reduced average incident-cause identification from 60 to 25 minutes.
Senior Software Engineer
Bloomberg LP
- Designed distributed fixed-income analytics processing 100M+ market and trade records daily for Bloomberg TOMS.
- Reduced analytics pipeline latency by 60% by migrating analytics calculations from SQL Server aggregations to Apache Druid.
- Designed the end-to-end pipeline for ingesting incoming trades, validating data, and calculating real-time metrics such as G-spread and Z-spread.
- Built back-end services and React Launchpad applications used by traders to interact with real-time analytics.
- Worked as part of a 5-person engineering team delivering and maintaining the TOMS analytics platform.
- Decommissioned 30+ legacy services, substantially reducing the legacy service footprint.
- Increased automated test coverage from 60% to 80%, a 20-percentage-point improvement, while maintaining production analytics for 100M+ daily records.
Software Engineer
Bloomberg LP
- Migrated legacy ASP.NET applications to .NET Core, improving screen/page-load performance by 70%.
- Reduced infrastructure costs by 30% through the .NET Core modernization.
- Modernized at least 40 application pages as part of the internal sales platform migration.
- Built a scikit-learn lead-recommendation system using call-note sentiment and logistic regression, increasing lead conversion by 12%.
- Achieved 90%+ positive feedback from sales teams for the lead-recommendation system.
- Built sales dashboards supporting 7 product lines and approximately 250 internal users.
- Developed dashboards serving 3 organizational levels, from sales representatives through executives.
- Automated sales-data aggregation using SSIS and ActiveMQ, reducing manual reporting effort by 50%.
- Saved approximately 1–2 hours per week for 180+ users through reporting automation.
- Built high-availability REST APIs and Bloomberg Terminal applications using C#, SQL, and JavaScript as part of an approximately 6-person engineering team.
Project History
Production Enterprise RAG Knowledge Platform
https://uplandsoftware.com/panviva/Enterprise Knowledge Platform Modernization and Microservices Migration
https://uplandsoftware.com/panviva/Distributed Fixed Income Analytics Platform
https://professional.bloomberg.com/products/trading/electronic-markets/fixed-income/AI Image Generation and Understanding for Text Documents
https://www.goodnotes.com/Education
Master's Degree in Information Technology
Rochester Institute of Technology - Rochester, NY, USA
Bachelor's Degree in Computer Science and Engineering
Manipal Institute of Technology - Manipal, KA, India
Certifications
Claude Certified Architect - Foundations
Anthropic
Skills
Libraries/APIs
React, REST APIs, Entity Framework, Node.js, Claude API, vLLM, Scikit-learn, Playwright
Tools
Claude, Slack, Microsoft Teams, Jira, Claude Code, Azure Kubernetes Service (AKS), Azure Application Insights, Jenkins, Grafana, Splunk, n8n, GitHub Copilot, Codex, Apache Druid, RabbitMQ, Tableau, Visual Studio, Claude Agent SDK, Azure App Service
Languages
TypeScript, Python, C#, JavaScript, SQL, HTML, CSS, C++
Frameworks
.NET, ASP.NET, AutoGen, .NET Core, Next.js, Agentic Frameworks, OAuth 2, Express.js, React Native
Paradigms
Model Context Protocol (MCP), Microservices, Automated Testing, Role-based Access Control (RBAC), Web Application Architecture, ETL, DevOps
Platforms
Docker, Azure, Azure Functions, New Relic, Visual Studio Code (VS Code), OpenStack, Azure AI Search, Bloomberg Terminal, Linux, MacOS, iOS, Ollama
Storage
PostgreSQL, MongoDB, Redis, Azure Cosmos DB, Databases, Microsoft SQL Server, JSON, Elasticsearch, Amazon S3 (AWS S3)
Other
Retrieval-augmented Generation (RAG), Embeddings from Language Models (ELMo), OpenAI GPT-4 API, CI/CD Pipelines, Gemini, SQL Server, Azure Service Bus, LLM Agents, Comdb2, OpenAI, Prompt Engineering, Azure Blob Storage, APIs, Agentic AI, Full-stack, Generative Artificial Intelligence (GenAI), Minimum Viable Product (MVP), Product Management, Microsoft Azure, OAuth, AI-generated Code, Architecture, Large Language Models (LLMs), Artificial Intelligence (AI), Software Architecture, AI Automation, ChatGPT API, Full-stack Development, API Integration, Azure Blog Storage, Back-end, role based access control, AI Agent Orchestration, AI Architecture, Workflow, Workflow Automation, Agentic Coding, API Design, Web Application Design, AI Agents, Agentic AI Systems, Observability, Agentic RAG Systems, Multi-tenant SaaS, Authentication, Third-party APIs, Cloud Infrastructure, Agentic Workflow Design, Cursor AI, Front-end, Caching, User Interface (UI), Workflow Automation & System Integration, Integration, AI Integration, AI-assisted Development, Bug Fixes, AI Engineering, FastAPI, Natural Language Processing (NLP), Machine Learning, Hermes, Fintech, Security, Cloud Security, Cloud Architecture, CRM, Integrations, ETL Pipelines, Monorepos, Cloud, Production Engineering, AI System Design, Claude Code Configuration, Context Management, Generative AI Architecture, Multi-agent Orchestration, Production Reliability, Structured Output, Tool & MCP Design, iOS App Development, Mobile App UI, Mobile App Development, Finance, Technical Leadership, ETL for AI
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring