About


Yuvraj Shiv

I'm an EECS undergrad at UC Berkeley, and a machine learning engineer co-op at Altametrics, where I'm prototyping computer vision for restaurants. Before the co-op I worked on production demand forecasting for three national restaurant chains, and with the other interns on the data pipeline behind it.

I like building the layer underneath the thing everyone else uses — the renderer, the query optimizer, the datapath, the data-collection rig. Most of what I've enjoyed has come from implementing something I'd only ever used as a black box and finding out it's tractable. The Scheme interpreter did that to me first; the RISC-V datapath did it again the next summer.

Outside of forecasting I co-founded Residual Robotics, where I owned the hardware for a wearable robot-manipulation data rig — the data bottleneck in manipulation is a hardware problem before it's a learning problem. I'm also drawn to database internals and computer graphics.

Experience


Altametrics

Machine Learning Engineer Co-op (Aug 2026 – present) and Machine Learning Intern · Costa Mesa, CA

  • Since Aug 2026, a full-time co-op on a computer vision prototype for restaurants, still in testing: YOLO object detection and multi-object tracking turn camera video into item counts, and fine-tuning a larger YOLO model raised precision, recall, IDF1, and count accuracy about 11% over the off-the-shelf model. LLM agent workflows check and label the processed data for next-action prediction.
  • Worked on production ML forecasting for daily sales across three national restaurant chains — Jack in the Box, Chipotle, and Buffalo Wild Wings — covering thousands of locations.
  • Cut forecast error from 12% to 5%, which translates directly into tighter labor scheduling and lower food cost.
  • Worked with the other interns on the data pipeline behind the model: scheduled crawlers behind a rotating proxy pool, a headless browser for pages that render client-side, and an ETL layer that lands everything in a cleaned, de-duplicated store.
  • Designed the schema and indexes — range partitioning by date and a composite B+ tree — and ordered the feature joins by selectivity to keep them fast.
  • Engineered 30+ predictive features from weather, zip-code demographics, holidays, seasonality, school calendars, and national events, plus macroeconomic indicators.
  • Worked on the full model lifecycle end to end: preprocessing, feature engineering, walk-forward cross-validation, hyperparameter tuning, and back-testing against historical sales.
  • Worked full-time in Summer 2025 and part-time in Summer 2026 — adding staffing forecasts that predict how many employees each time slot needs — alongside summer-session coursework (CS 61C in 2025, CS 184 and CS 161 in 2026).
Python · XGBoost · LightGBM · CatBoost · Prophet · TimesFM
2025 – present

Residual Robotics

Co-founder · hardware · Berkeley

  • Co-founded a 4-person startup that planned to collect mass UMI robot-manipulation data from skilled workers in India, to sell to robotics foundation model labs.
  • Owned the hardware: designed 3 of the 16 parts in the handheld gripper's parametric Onshape assembly, and designed a custom PCB around a Raspberry Pi Zero for the sensing stack.
  • Cut the CAD model from 636 g to 200 g through topology optimization, and did the design-for-manufacturing work to print all 16 parts with zero slicer supports.
  • Raised $10K in pre-seed funding and was invited to an in-person interview at Y Combinator, which accepts under 1% of applicants.
Onshape · parametric CAD · topology optimization · DFM · Raspberry Pi Zero
Jul – Oct 2026

MAG Services

Software Engineering Intern · Costa Mesa, CA

  • Built Java data-analysis tooling over 12+ weeks of restaurant sales and labor-cost data, surfacing scheduling inefficiencies across 20+ locations.
  • Designed a labor-optimization playbook adopted by operations managers that cut overtime and eliminated unprofitable shifts.
  • Delivered 15+ analytical reports and stakeholder presentations across 20+ recurring meetings.
Java · SQL · Excel
May – Aug 2023

COSMOS — UC Santa Cruz

Semiconductor Research · Baskin School of Engineering

  • Selected for COSMOS Cluster 10: Semiconductor Materials and Device Engineering, a competitive UC summer research program; worked with solid-state device physics: bandgaps, charge carriers, pn junctions, and the operation of LEDs, transistors, solar cells, and memristors, plus semiconductor synthesis and device/IC fabrication.
  • Led a 4-person team to design and prototype a 4-watt solar panel.
  • Co-authored and presented a technical project on memristor applications in neuromorphic computing and ML hardware acceleration.
Jul – Aug 2023

Education


University of California, Berkeley

B.S. Electrical Engineering & Computer Science

Coursework across systems, ML, and EE — see the coursework page for the full list and what each course actually involved.

Foundations of Computer Graphics (CS 184)
Computer Security (CS 161)
Database Systems (CS 186)
Internet Architecture & Protocols (CS 168)
Artificial Intelligence (CS 188)
Discrete Mathematics & Probability Theory (CS 70)
Computer Architecture (CS 61C)
Data Structures (CS 61B)
Structure & Interpretation of Computer Programs (CS 61A)
Designing Information Devices & Systems I (EECS 16A)
Introduction to Circuits & Devices (EECS 16B)
Aug 2024 – 2028

Riverside City College / Norco College

Associate's degree · dual & concurrent enrollment

Earned an associate's degree concurrently with high school, at a 4.0 GPA. I entered Berkeley with junior standing on day one, from 70+ transferable units earned during high school.

Aug 2021 – May 2024

Activities


Cal Bhangra

Outreach Coordinator, 2025–26 season

Cal Bhangra is UC Berkeley's competitive Punjabi folk dance team.

2025 – 2026

AP Calculus AB/BC Head Teaching Assistant

Hillcrest High School · Riverside, CA

Supported instruction across two AP Calculus sections: graded coursework for 145 students, authored solution sets for 20+ unit tests, and ran the course's Google Classroom of lecture notes and review material.

2022 – 2023

American Red Cross — Southern California

Volunteer

Installed 1,000+ smoke alarms across 250 homes through the Sound the Alarm program, helped coordinate an ARC Expo for 200 attendees, and supported multiple blood drives.

2022 – Present