Large Language Models: architecture, pre-training, and alignment safety.
I am a deep learning engineer focused on optimizing model capability, safety frameworks, and scaling
efficiency. I research transformer pre-training architectures and develop safety-critical alignment
frameworks to prevent prompt injections and tool exploitation.
See my LLM knowledge in this repo ↗
Core Areas
Pre-training & Adaptation
Pre-training medical and domain-specific transformers using SwiGLU, Grouped Query Attention, and
Rotary Embeddings. Architecting memory-efficient adaptations using LoRA/QLoRA on cloud clusters.
Alignment & Model Safety
Implementing alignment feedback loops (SFT, DPO, GRPO) for safety and instruction compliance.
Building multimodal input/output guardrails for VLM applications — chain-of-thought safety
classification served via vLLM, with safety policy diagnostics across risk categories.
Experience & Education
Chronology of professional internships and academic coursework.
May 2026 — August 2026
A10 Networks
San Jose, CA
Deep Learning Engineer Intern — AI Firewall Team
- Architected and trained TrooperViz, a multimodal safety guardrail model built with Python, PyTorch, and Hugging Face Transformers, enabling comprehensive evaluation of text, image-text inputs, multi-turn conversations, and assistant-generated outputs across diverse safety categories.
- Fine-tuned a vision-language model for multimodal safety classification and merged it with a text-only guardrail model using SLERP, creating a unified safety system with broader coverage across visual, conversational, and generative-AI interactions.
- Engineered end-to-end data and evaluation workflows—including annotation, Pandas-based preprocessing, synthetic multi-turn conversation generation, model benchmarking, and interactive inference via Gradio and vLLM—streamlining the testing pipeline and accelerating safety assessments.
Python · PyTorch · SFT · SLERP · vLLM · Hugging Face · W&B · CI/CD
Oct 2025 — Present
Routes Technologies
Remote, TX
AI Engineer Intern
- Architected an LLM-powered NL-to-SQL RAG pipeline with few-shot prompting, input normalization (470+ synonym mappings), parameterized SQL sanitization, session memory, and Weights & Biases observability—deployed across 5 Azure ML managed online endpoints.
- Built an SVD collaborative-filtering recommendation pipeline on Azure ML, evaluated with precision, recall, and MAP metrics, and paired it with an LLM-driven taste profiler that batch-generates user preference vectors in a nightly ETL job, resulting in more relevant recommendations and faster cold-start handling.
- Engineered a multi-source recipe ingestion system using a Scrapy crawler with JSON-LD extraction, Instagram Graph API and TikTok oEmbed integrations, GPT-4o-mini for recipe classification, lingua for language filtering, and faster-whisper for audio transcription fallback, which expanded the recipe catalog and cut ingestion latency.
Python · SQL · Azure ML · scikit-learn · Sentence-Transformers · W&B · Scrapy
May 2025 — Aug 2025
Dreamable Inc.
San Francisco, CA
AI/ML Engineering Intern
- Trained, tuned, and deployed ML models in a production Cloud environment using Python, PyTorch, TensorFlow, and Hugging Face, working collaboratively with a cross-functional team.
- Fine-tuned Qwen-2.5-7B via LoRA on GCP Cloud to lower training cost while delivering a Q&A model with comparable accuracy within budget, tracking experiments with Weights & Biases.
- Curated NLP datasets (Pandas, NumPy, Hugging Face Datasets) to improve training data quality, and built a generative-AI outreach agent with LangChain and the OpenAI API to automate messaging workflows and increase response rates.
Python · PyTorch · TensorFlow · Hugging Face · LoRA · GCP · LangChain · W&B
Education
San Jose State University
2026 — 2027
B.S. Computer Science · GPA 3.88
- NSP research with Professor William
- CodePath Advanced DSA training
- Dean's List Honoree
San Francisco State University
2023 — 2025
Computer Science (Transferred)
- VP of AI Club & Tech Lead at SparkSF
- Hosted SFHacks (400+ attendees)
- Dean's List Honoree
Projects & Models
Pre-trained models, adapters, and open-source repositories.
Repositories
LLM Firewall for Agentic Tool-Calling
Low-latency inline defense intercepting prompt injections. GPT-2 attacker loop reduced bypass from 23.1% → 3.72%. BERT + LoRA adds ~20ms latency.
GitHub ↗
theHelper — AI Research Assistant
Production RAG with FAISS, LangChain chunking, cross-encoder reranking. Local observability into daily JSONL. CI-gated QA.
GitHub ↗
Kanting — Video RAG System
Indexes YouTube transcripts via Whisper. Semantic search across Sentence-Transformers DB returning precise clip timestamps.
GitHub ↗
AdaptRoute
Task aware routing based on the query to task specific models
GatorGPT
63M transformer for consumer GPUs. Grouped Query Attention and RoPE.
How LLMs Are Made
Annotated code building GPT-2, DeepSeek MoE, and Kimi from scratch.
GitHub ↗
Hugging Face Weights
MedAssistGPT 303M & 401M
Medical-domain transformers on PubMed. SwiGLU, GQA, RoPE.
HuggingFace ↗
Qwen2.5-0.5B SFT+DPO 0.5B
Chat model fine-tuned with SFT and Direct Preference Optimization.
HuggingFace ↗
Llama-3.2-3B OpenHermes 3B
QLoRA on filtered OpenHermes conversational datasets.
HuggingFace ↗
StableLM 1.6B SFT+GRPO 1.6B
Aligned via GRPO on PKU safety preferences.
HuggingFace ↗
Skills & Credentials
Technical expertise, hackathons, and certifications.
LLM Engineering
Transformers · SFT · DPO · GRPO · PPO · LoRA/PEFT · TRL · vLLM · Quantization · Vector DBs · Prompt Eng.
ML & NLP
PyTorch · TensorFlow · Scikit-learn · LangChain · FAISS · Whisper · BART · Sentence Transformers · Pandas · NumPy
Backend & Cloud
FastAPI · Flask · Docker · Azure ML · GCP · PostgreSQL · MongoDB · Scrapy · Nginx · CI/CD
Programming
Python · SQL · Java · JavaScript · C++ · Bash · R · HTML/CSS · Git · Linux
Hackathons
CalHacks 12.0 — Palace of Fine Arts, SFOct 2025
MCP AWS Agentic Challenge — AWS Builder Loft, SFJul 2025
SacHacks — VirtualMar 2025
HackMerced — UC MercedMar 2025
Cal Hacks 11.0 — San FranciscoOct 2024
Certificates
- AI Memory: LLM Memory Systems — LinkedIn
- Fine-Tuning for LLMs: Beginner to Advanced — LinkedIn
- Model Context Protocol (MCP) — LinkedIn
- Introduction to Generative AI — Google Cloud
- Introduction to Web Development — UC Davis
- Programming in Python — University of Michigan
- Special Theory of Relativity — Stanford University
- Calculus through Data & Modelling (×4) — Johns Hopkins
Get in Touch
Send a brief message to open collaboration.
Open to inquiries about custom fine-tuning runs, alignment evaluation, and model safety audits.