Marc de Lecea

Experience

Lab Robotics Engineer III

Robotic Systems & Workcells

• Designed and built two dual-robot-arm workcells from scratch, delivering lights-out scale-up at a fraction of the time and cost of commercial integrators.
• Developed a Python constraints-based scheduling algorithm for in-house robotic workcell orchestration, improving throughput and reducing integration cost.
• Successfully onboarded a fully integrated HighRes system, increasing automation bandwidth by 600%.
• Designed an Arduino-based precision dosing system with closed-loop PID control for continuous reagent delivery.
• Engineered biosafety hardware — HEPA-filtered enclosures, fume extraction shields, and automated tip-waste systems — to meet lab safety requirements.

Liquid Handling & Lab Automation

• Created an LLM-powered platemap transformer for users to quickly create complex plate layouts, paired with a modular Python API that transforms platemaps and generates worklists for liquid handlers.
• Developed a universal Lynx liquid handler worklist parser that optimizes multi-dispensing, splits volumes to fit tips, and automatically arranges labware to enhance operational efficiency.
• Designed intelligent Lunatic transfer and read methods, enabling fully automatic normalization from concentration readings by updating the database and calling a custom API.
• Elevated barcode coverage from 0% to 70% of lab processes through new tools and trainings.

Software & Data Infrastructure

• Designed and implemented an API to receive and parse real-time usage data from lab instruments, ensuring prompt alerts through Slack messages if an instrument requires attention.
• Hosted and integrated a Streamlit data dashboard using live lab instrument data from the API, enhancing data visualization and analysis capabilities; Dockerized services for deployment on a Kubernetes cluster.
• Collaborated with the Informatics team to host the API using AWS Lambda and implemented an Aurelia/TypeScript frontend for the tools.
• Improved liquid handler integration driver reliability and reduced command latency by 1000% by redesigning the driver layer around a real-time push-queue architecture.

Machine Learning & Perception

• Extended the Lynx parser with a supervised-learning deck placement system that predicts scientists' preferred labware layouts from operator corrections, paired with a collision-detection engine modeling labware, tip heights, and head offsets to catch physically invalid configurations before execution — reducing operator error and setup time by roughly 65%.
• Built an OpenCV-based computer vision system for deck and labware validation, catching missing or misplaced labware before protocol execution.
• Prototyping a LeRobot-based tube-handling robot, applying imitation learning policies toward autonomous pick-and-place manipulation.

Leadership & Mentorship

• Effectively mentored and guided junior engineers, fostering a collaborative and supportive team environment.

Manufacturing Automation Engineer

• Reverse-engineered an undocumented communication protocol for a laser cutter with no vendor API, then built a custom Python driver to dynamically generate G-code and spoof execution commands — replacing a fully manual process (operators hand-writing serial numbers with a sharpie) with automated laser engraving. The result: batches of ~120 units engraved in 2 minutes, with more legible, wash-resistant, higher-capacity serial codes.
• Designed a SolidWorks hardware fixture to improve operator ergonomics; authored usage, maintenance, and validation documentation for manufacturing.
• Prototyped and built a batch PCB cleaning tool, speeding up flux cleaning by 1500%.
• Modified Android testing software to speed up probe verification and increase accuracy.

Research Assistant

• Applied A* pathfinding algorithm and plane mapping to microrobot swarm simulation to provide control for previous work and broaden robot capability.
• Created MATLAB GUI to assist visualization of microrobots and their path planning decisions.
• Used LabVIEW to create manual, joystick-based control schema of microrobots using magnetic field array.
• Implemented vision tracking system to measure and automatically control multiple microrobots simultaneously.
• Crafted automatic control and calibration system to accurately direct robots toward desired destination.

Machine Design Teaching Assistant

• Created interactive demo SolidWorks models and python scripts for students studying at home to demonstrate movement and fundamental concepts.
• Led office hours and published short discussion videos reviewing principle equations and problems.

Fulfillment Associate

• Optimized workflow by increasing accessibility to product depots.
• Ensured customer satisfaction with timely and thorough item searches.
• Leveraged company database to fullest extent to assess product placement.

Engineering Intern

• Developed novel root imaging robot in collaboration with PhD student to determine plant health and yield.
• Used ROS to simulate and program robotic arm.
• Used SolidWorks to develop alternative mechanical solutions, saving cost and complexity.

Engineering Intern

• Created learning materials for "Bring your child to work day", including a Snap! programmable car and blockchain demonstration using Micro:Bits.
• Tinkered with Android Things and AWS DeepLens to expand platform availability for team and exhibit new avenues of approach.

Engineering Intern

• Designed and machined parts for a lab-grade high-throughput microscope.
• Given design objectives, planned and designed parts, and got parts approved by team of five professional engineers.
• Used SolidWorks to CAD lens/mirrormounts, 3D-printed mockups, custom crates, I/O panels, and badge jigs.

Intern

• Modeled 3D views of experimental pig hearts to be printed out of agar compound using a MATLAB GUI.