Experience

Knightscope Inc.

Jan 2021 – Present

Sr. Machine Learning Engineer · Mountain View, CA

Sr. Machine Learning Engineer Jan 2025 – Present

Own the AI/CV stack within the K7 ICM (Intelligence Control Module) on Knightscope’s next-generation autonomous security robot — including object detection (person, face, ALPR, thermal, vehicles) and live thermal streaming — developed to NIST 800-53 security controls.
Designed and implemented the consolidation of 3–4 disparate AI runtime pipelines into a single cross-platform stack on Nvidia Jetson Xavier, reducing duplicated maintenance work across product lines and lowering DevOps overhead.

Machine Learning Engineer Jan 2021 – Dec 2024

Sole ML engineer owning Knightscope’s end-to-end AI stack across 3 product lines and 100+ deployed robots serving 50+ clients nationwide, with per-robot subscriptions ranging $30K–$100K annually.
Supported Knightscope’s 2022 IPO (3PAO audit, FEDRAMP authorization) and subsequent Federal ATO (Authority-to-Operate) by porting AI inference workloads from Jetson edge devices onto a FIPS-eligible x86 platform, enabling the first Gov-Cloud K5v5.2 deployment at the U.S. Department of Veterans Affairs (Texas, 2024).
Upgraded the AI software stack for the K5v5.1 robot inaugurated at NYPD by Mayor Eric Adams (2023) — Knightscope’s first NYC city-government partnership.
Migrated 100+ deployed robots from Ubuntu 16.04 on Jetson TX1 to Ubuntu 18.04 on Jetson Xavier AGX, introducing virtualenv-based isolation and Ansible-driven OTA release workflows.
Migrated cloud inference from legacy Ubuntu 14.04 EC2 to 22.04 g4dn.xlarge behind an Nginx + uWSGI + Flask stack with self-healing workers, serving up to 10 req/s across the fleet.
Optimized real-time edge detection across the deployed fleet by quantizing object detection models to INT8/FP16 via TensorRT and adding a per-detector ByteTrack post-processing layer (Kalman filter with two-round association) — reducing cellular costs by 20% and network bandwidth by 50%.
Authored a sales-process audit flagging gaps in client-facing feature representation, recommending fixes aimed at reducing churn on multi-year subscriptions.

Squark

Oct 2020 - Dec 2020

Machine Learning Engineer

Worked on developing a system to observe and visualize Bias and ensure Fairness in machine learning systems using Statistical Parity Difference, Equal Opportunity Difference, Average Odds Difference, Disparate Impact and Theil Index Metrics.
Mitigating Bias and ensuring fairness using optimized pre-processing & adversarial debiasing algorithms.
Also working on reducing data dimension using t-Stochastic Neighbor Embedding (t-SNE) and developing interpretable models using Local Interpretable Model Agnostic Explanations (LIME).

UII America Inc.

Sept 2019 - Jan 2020

Computer Vision Research Intern

Implemented Dense Correspondence Estimation algorithm for real-time patient inference and increased inference speed by 50% to ~20-21 FPS using multi-threaded Dockerized architecture on Nvidia P5000 GPU
Developed a technique to acquire skeletal positions of human body by reading depth image frame from RGBD images captured using Astra Orbbec Depth Camera.
Annotated image data using Densepose for keypoint estimation and ran the system on NVidia P5000 GPUs.

Northeastern University

Jan 2019 - Apr 2020

Graduate Teaching and Research Assistant

Worked on Mouse Brain Image Analysis by trying to classify sagittal, horizontal and coronal views of images. Used ResNet-50 architecture for training the models on 2 NVidia Tesla T4 GPUs.
Reduced run-time complexity by 60% by developing micro-services architecture using H2O auto-modeling framework.
Memory and time profiled metrics on Tensorboard using PyTorch and achieved 2x faster data-loading with Dask.
Assisted Dr. Nik Bear Brown in his course of INFO 7390 - Advances in Data Science and Architectures and INFO 6105 - Data Science Engineering Tools and Methods.
Conducted Deep Learning workshops on Convolutional Neural Networks for batch of 100+ students. Also managed a batch of 50 students and assisted them in the academic course.

InVideo

Sept 2017 - May 2018

Software Engineer

Scaled up video production rate for clients from 30 videos/day to 300 videos/day by implementing Geometric Perspective Transformation on image masks
Improved software efficiency by 60% by saving 10 mins of rendering time per video by incorporating Face Detection using Convolutional Neural Networks for automated text positioning
Parsed news articles for understanding language semantics using Spacy and NLTK for text-to-video translation
Integrated additional features for our product like providing users the ability to add GIFs, social media widgets, image icons, etc for contextual representation of information in videos.

K. J. Somaiya College of Engineering

May 2017 - Aug 2017

Research Assistant

Worked Under Dr. Jagannath Nirmal, HOD, Electronics Department, KJSCE on industrial Machine Learning and Computer Vision applications.
Worked on a project demanding the need to develop a system to classify Dysphonia patients from normal patients using Machine Learning. It uses an algorithm which uses a robust feature extraction technique named I-Vectors for classification. Entire implementation done in MATLAB and can be migrated into Python packages. Successfully modelled a system which gives 98% accuracy using the proposed technique as compared to the 92% before using the technique. Used Support Vector Machines (SVM) to perform Binary classification.¹
Simultaneously worked on another project requiring the need to classify patients with Dysphonia, Paralysis, Laryngitis and Normal. Used the above proposed technique of I-Vectors to develop 3 different models to classify these 4 classes. Support Vector Machines (SVM), Naive Bayes and K-Nearest Neighbours (KNN) were used with multi-class functionality. The entire system was developed in MATLAB and can be migrated into Python packages.²

Nikunj Rajesh Lad