Anudeep Hegde
Building AI systems in production — not demos.
Backend & Applied AI engineer focused on LLM systems, RAG, evaluation, automation, and reliable cloud-native deployment.
Some of My Favorite Works
Systems deployed across investment, operations, and engineering teams
Deep Reinforcement Learning Portfolio Optimization System
2025 Innovation AwardBuilt and productionized AI workflows for portfolio decision-support with Quant Research and Data Engineering.
Enterprise GenAI Operations & Analysis Platform
Enterprise ScaleEnd-to-end system with backend connected to OpenAI LLM, powering investment operations with news analysis, document review, compliance workflows, and sentiment analysis.
Automated Scenario Testing & Regression Agent
Company-wide Production DeploymentBuilt a Python backend and GenAI pipeline that extracted JIRA criteria, generated QA test cases, and integrated with qTest.
About Me
I studied Computer Science at UC Irvine with a focus on AI/ML, and I'm currently working through UC Berkeley's Master of Data Science program. These days I spend most of my time building AI systems that get used by real teams — investment analysts, operations staff, QA engineers.
Day-to-day, that means FastAPI backends, RAG pipelines, LLM integrations, and figuring out how to make these things reliable at scale. I've spent a lot of time on evaluation and monitoring because I've learned the hard way that an AI system is only as good as your ability to measure it.
I care about systems that are reliable, measurable, and useful.
Outside of work, you'll find me experimenting in the kitchen, at the gym, or planning trips with friends.
Technical Skills
The tools and technologies I work with daily
Languages
Backend Systems
Applied AI / LLM Systems
Cloud & Infrastructure
Experience
Building enterprise-grade solutions at scale
Capital Group
CurrentSoftware Engineer
Building and productionizing applied AI systems for investment and compliance workflows.
Capital Group
Software Engineering Intern
Developed a GenAI pipeline for automated test case generation, reducing manual effort by 86%.
Microsoft
Student Tech Program Co-op
Built serverless Azure workflows and ML-powered services to improve automation and prediction accuracy.