Backend Systems
APIs, orchestration layers, data pipelines, persistence, and service architecture.
Software Engineer at Meta
Building backend, infrastructure, and AI-native product systems.
I'm Peng Zhang, a software engineer focused on backend and platform problems, product-facing systems, and developer tooling. My work spans API design, orchestration, retrieval pipelines, streaming interfaces, and applied AI systems that need to work reliably in practice.
I bring a builder's mindset with research depth: I like shipping useful systems, reducing complexity, and staying close to fast-moving areas like LLM applications, foundation models, and trustworthy ML.
Current
Software Engineer, Meta
Focus
Infra, Product, Foundation Models
Background
Systems building + applied ML research
Focus
What I build
Developer Tools
Workflow automation, code intelligence, review systems, and tools that improve engineering leverage.
AI Applications
LLM-powered products, agent pipelines, retrieval systems, and pragmatic model-driven UX.
Profile
How I work
Systems Mindset
I like clean interfaces, clear abstractions, and systems that stay understandable as they grow.
Product Taste
I care about building things people can actually use, not just technically interesting demos.
Research Fluency
I can move between implementation and research contexts, especially around LLMs, retrieval, and ML systems.
Stack
Tech Stack
C++
Java
Python
PHP
TypeScript
Linux
FastAPI
Next.js
React
SQL
Docker
Kubernetes
LLM Systems
ML Infrastructure
Selected Work
Featured projects
ResearchPilot
A full-stack research copilot that retrieves papers, extracts structured findings, synthesizes literature, and drafts citation-aware related work through a live, streaming workflow.
- Built a production-style architecture with FastAPI, Next.js, SSE streaming, report history, and semantic search.
- Designed an agent pipeline across retrieval, extraction, synthesis, and writing with persistent storage and vector search.
RepoReviewer
A local-first code review system for public GitHub repositories and pull requests, built around a multi-agent review pipeline and a shared backend for both CLI and web workflows.
- Combined CLI and web interfaces around a reusable review engine with FastAPI APIs and live status updates.
- Added prioritization, artifact generation, and file filtering for practical large-repository review workflows.
CLARA Medical Dialogue Simulation
A prototype dialogue simulation system for history taking and diagnostic reasoning, supporting both deterministic and LLM-enhanced analysis paths.
- Built a modular pipeline spanning transcription, semantic alignment, feedback generation, and performance analytics.
- Designed graceful fallback from LLM-enhanced flows to deterministic execution paths.
Research
Publications
2025
Hybrid Deep Learning Framework for Enhanced Melanoma Detection
IEEE Transactions on Computational Biology and Bioinformatics, 2025
Leveraging Clinical Record Geolocation for Improved Alzheimer's Disease Diagnosis Using DMV Framework
Biomedicines, 2025
Beyond Detection: Predicting Code-Switch Points in Multilingual Conversations
WiML Workshop at NeurIPS, 2025
When Pain Hides in Silence: ML-Driven Flare-Up Prediction for Hidradenitis Suppurativa Using Synthetic Patient Data
GenAI4HealthInfo at IJCAI, 2025
Leveraging Geolocation in Clinical Records to Improve Alzheimer's Disease Diagnosis Using DMV Framework
arXiv preprint, 2025
2024
Hybrid Deep Learning Framework for Enhanced Melanoma Detection
BIOKDD, 2024
Experience
Professional timeline
2025 - Present
Software Engineer
Working across infrastructure, product and ecosystem systems, and ML foundation model initiatives.
2024 - 2025
Research Assistant
Worked on applied machine learning and research engineering projects.
2024 - 2025
Research Assistant
Contributed to research around generative AI, NLP, and multilingual systems.
2024 - 2025
Teaching Assistant
Courses: CS6240 Large-Scale Parallel Data Processing and CS6140 Machine Learning.
2024 - 2025
Peer Mentor
Mentored students in the MSc in Computer Science Align program.
2024 - 2025
Student Chair
Helped coordinate workshop operations for a research venue focused on LLMs and generative AI.
2020 - 2022
Software Engineer
Worked on dashboard platform infrastructure and data pipeline systems.
2018
Founder (of SCOPE Camp)
Education
Academic background
Northeastern University
2022 - 2024
Khoury College of Computer Sciences
MS in Computer Science
Shanghai University
2016 - 2020
BS in Business and Information System