UC Berkeley EECS

Daniel Firoozabadi

Daniel Firoozabadi
Berkeley, CAEECS

Work

Experience

UC Berkeley Sky Computing Lab

Undergraduate Researcher

Jul 2026 — Present Now

  • SGLang
  • LLM inference
  • Scheduling
  • GCP A100
  • Extending Dev Bali's FairInference project (SOSP 2026 submission), implementing scheduling logic in a forked SGLang inference engine using a spec-driven development approach with defined pre/postconditions.
  • Designing a fairness policy for hybrid prefill/decode batching, benchmarking against baseline policies on a GCP A100 GPU while implementing parallelization to bound latency interference across concurrent clients.

Capital One

Software Engineering Intern

Jun 2026 — Present Now

  • LangGraph
  • Chroma
  • PyArrow
  • Kubernetes
  • Cut Excel reconciliation from 4 weeks to under 2 days by building an agentic RAG system with LangGraph, Chroma, and PyArrow for a $10B-spend ledger migration.
  • Returned discrepancy lookups in under 3 seconds across 6,500+ SOX accounts via hybrid graph and semantic retrieval over 6 data sources and 6M+ rows.
  • Enabled 20+ Finance analysts to self-serve reconciliation by deploying a containerized ReAct agent with Docker, Kubernetes, an 8-tool toolkit, and a Gradio interface.

Stanford Graduate School of Business

Software Engineer — AI Security (via Ascend Berkeley)

Jan 2026 — May 2026

  • Multi-agent
  • AutoGen
  • PentAGI
  • Nmap
  • Nuclei
  • Led architecture planning and built an end-to-end multi-agent penetration testing system coordinating reconnaissance, vulnerability classification, validation, and report generation with AutoGen, PentAGI, Nmap, and Nuclei.
  • Implemented human-approval gates, target allowlists, append-only audit logging, and an emergency kill switch; confirmed 5 valid vulnerabilities from approximately 10 findings in controlled environments.

Olfera

Software Engineering Intern

May 2025 — Aug 2025

  • AWS Amplify
  • MySQL
  • GraphQL
  • Built and deployed backend infrastructure on AWS Amplify, integrating authentication and CI/CD pipelines across development, staging, and production environments.
  • Designed 7+ normalized MySQL schemas and secure GraphQL/AppSync endpoints with role-based access control, reducing redundant queries by approximately 40%.

Runopt

Software Engineering Intern

Jan 2025 — May 2025

  • PyVista
  • OpenMDAO
  • Developed a PyVista 3D visualization framework for civil-site design, enabling engineers to simulate stormwater detention layouts in under 10 seconds per run.
  • Implemented a knapsack-based optimization pipeline across 9+ UGDC models and integrated OpenMDAO, reducing cross-module runtime by approximately 30%.

USC Flying Light Specks Lab

Research Intern

May 2024 — Aug 2024

  • Arduino
  • Controls
  • Autonomous flight
  • Programmed autonomous flight behaviors for 5+ micro-drones using Arduino-based control of pitch, roll, and yaw, applying physics and controls principles to develop stable 3D flight patterns.
  • Designated as a contributor to the NSF award of $250,000 for the project "EAGER: Prototyping Touch with FLS-Matter."

Projects

Things I've shipped

Tandem

TypeScript · React · Node.js · Anthropic SDK · WebSockets

  • Built a real-time collaborative coding environment enabling multiple developers to share, branch, and replay live AI sessions simultaneously.
  • Advanced to the final partner interview for a16z Speedrun after developing the prototype at the Stanford x Google DeepMind Hackathon.
Demo login
Emailtandemdemo2026@gmail.com
PasswordWelcomeTandem2026!

ShopSync

Next.js · Supabase · Vercel · OpenAI API

  • Built a full-stack LLM recommendation platform that extracted and normalized inventory data to generate personalized product matches and business listing suggestions.
  • Implemented ingestion and embeddings-based retrieval pipelines over scraped and API-sourced product data in Supabase; selected as a Falling Walls Lab SF finalist.

Goldman Sachs Mutual Fund Challenge

Java · Spring Boot · Angular · OpenAI API

  • Built for the Goldman Sachs Emerging Leaders 2026 challenge — a portfolio projection tool combining CAPM expected-return modeling with LLM-generated allocation.
  • Connected the Angular frontend to the Spring Boot backend's future-value projection endpoint as part of a 5-person team.

Goldman Sachs Emerging Leaders 2026 · team of 5

Education

UC Berkeley

B.S. Electrical Engineering & Computer Sciences · GPA 3.5/4.0

Data Structures & Algorithms · Computer Architecture · Optimization Models · Discrete Math & Probability

Selected for

Goldman Sachs Emerging Leaders Series — Software Engineering Track, Dec 2025 – May 2026. Built quantitative analytics infrastructure for a mutual fund forecasting platform.

Ascend Berkeley (Tech Chair) · AI Entrepreneurs at Berkeley · IEEE · YC Startup School 2026 · 2× SIG Discovery Day · Point72 Academy Spring Sessions

Stack

Languages

Python, Java, C, C++, SQL, JavaScript, TypeScript

Frameworks & ML

PyTorch, TensorFlow, NumPy, Pandas, React, Node.js, LangGraph, LangChain, AutoGen

Tools

Linux/Unix, AWS (Amplify, DynamoDB), Docker, Kubernetes, Git, MySQL, PostgreSQL, Supabase, CI/CD, REST APIs, GraphQL, Terraform

Contact

dfiroozabadi@berkeley.edu
Berkeley, CA GitHub · LinkedIn