experience
Research, software engineering, and data engineering roles.
Garner Health
Software Engineer Intern current
- Engineered a full-stack observability pipeline with automated dashboard and alert provisioning, using distributed tracing across Datadog RUM/Replay and OpenTelemetry (Loki, Prometheus, Tempo) to correlate frontend sessions with backend traces.
- Architected an autonomous on-call triage AI agent that ingests cross-stack telemetry to run root-cause analysis and generate verifiable fix pull requests under zero-PHI guardrails, resolving 10+ production incidents.
Columbia Data, Agents, and Processes Lab
Data Agents Research Assistant current
- Optimized core evaluation infrastructure for question-answering agents over a 9.5 TB/~40M-document data lake, including parallel sandboxed benchmark workers, tool-call and reasoning telemetry, and BM25 and hybrid search engines.
- Improved GPT-5-mini semantic match from 2.22% to 56.3% by adding context compaction, loop-detection plugins, structured search-result context, and stronger data-analysis tools.
- Built an ablation framework that swaps in idealized search, planning, and data-analysis tools to isolate where agents fail on grounded question-answering over long context windows.
Columbia Zuckerman Institute
Machine Learning Research Assistant
- Compressed Qwen2-Audio's 4096-D hidden states into 10-100D feature-specific representations, each preserving one linguistic or paralinguistic feature — such as emotion or reasoning — while suppressing the others.
- Curated training-ready datasets from 12 speech and language benchmarks, including converting text benchmarks into audio through parallel Kokoro-TTS generation.
- Built and parallelized a GPU-optimized PyTorch pipeline for Qwen2-Audio inference and hidden-state extraction, bottleneck-width sweeps, and supervised classification under cross-validation across 12 speech attributes and 120K+ audios.
Ruangguru
Data Engineering Intern
- Consolidated millions of payment events into user sessions in BigQuery, producing a product funnel used to identify major checkout drop-off points.
- Automated Google Review scraping and sentiment analysis with Puppeteer and Hugging Face; processed tutor-scheduling data into live Looker Studio heatmaps for operations teams.