Recruiter brief
Vraj Patel
Systems integration engineer who turns operational friction into reliable software.
I build internal software where reliability, data quality, and operator judgment matter. At CU Boulder’s Institute of Behavioral Science, I work across backend services, operational analytics, automation, and ML-assisted workflows.
At a glance
- Current role
- Systems Integration Engineer
- Location
- Boulder, Colorado
- Education
- Dual M.S. · Computer Science + Engineering Management
- Strongest fit
- Systems Integration · Backend/Platform · Data Engineering · Applied ML Infrastructure
Selected evidence
Results and scale
- 82.6% / 0.756
- Sealed-test accuracy / macro-F1 for an unreleased BERT routing candidate
- 10.8K / 96.7%
- Current ticket corpus / records carrying support-group IDs
- 61% / 66% smaller
- Raw / gzip analytics payload after replacing Plotly with Recharts
- ~15 min → <1 min
- Typical Active Directory group cleanup after PowerShell automation
Engineering profile
Backend, data, and applied ML systems
I work on internal tools where source data is imperfect, jobs can run for hours, and a person still needs to review the result. I build the state, checkpoints, and approval boundaries that keep those workflows usable.
I have applied the same approach to public-interest data systems, NLP experiments, and anomaly detection on synthetic satellite telemetry through Purdue’s Data Mine and an L3Harris collaboration.
I am completing dual M.S. programs in Computer Science and Engineering Management at CU Boulder, where I earned a B.S. in Computer Science with a Business minor.
Selected proof
Three projects to inspect
- 01Operational Ticket Intelligence
Full-stack support operations over a current 10.8K-ticket corpus, with 96.7% carrying support-group IDs, read-only reconciliation, and sealed-test model evaluation.
- 02Indian Parliamentary Data
Recoverable crawling, transcription, translation, PDF text extraction, and document normalization for multi-decade public-interest research.
- 03SeeMyRace
Full-stack race-photo retrieval system built primarily on open-source frameworks; I shipped verification, GPX race creation, admin, and upload/search workflows across 32 mainline commits, with work landing through seven merged PRs.