Pranay Reddy Samala

Product Engineer

Stanford, California, United States3 yrs 6 mos experience

Most Likely To SwitchAI ML Practitioner

Key Highlights

Expert in machine learning and algorithm development.
Pioneered privacy-preserving SLAM technology.
Developed safety models for autonomous driving.

Stackforce AI infers this person is a Machine Learning Engineer with expertise in Robotics and Autonomous Systems.

Contact

Skills

Core Skills

Machine LearningComputer ScienceArtificial Intelligence (ai)Algorithmic TradingSignal Processing

Other Skills

Visual SLAM algorithmsCamera-only localizationModel trainingTuningQuantizationCamera sensor experimentationBehavior prediction modelsDriving behavior analysisMachine learning methodsSimulation techniquesParticle solversMachine learning pipelineTime series modelsCounterfactual evaluationMachine learning strategies

About

I enjoy building hard tech from the ground up, especially at the intersection of systems and ML. I’ve worked across early-stage startups and large-scale autonomy teams, with a focus on pushing the frontier of what can be achieved. My work includes: • Parallel Web Systems – 8th in the company: Highlights include a one-of-a-kind web monitoring system, state-of-the-art web task APIs and deep research workflows. • Matic Robots – Core focus on the camera-only localization system. Pushed it into one of the best on-device solutions in the market, working end-to-end across model training, tuning, quantization, and camera sensor experimentation. • Waymo – Trained behavior prediction models focused on capturing rare and high-risk driving behaviors, including the tail end of bad driving scenarios.

Experience

3 yrs 6 mos

Total Experience

1 yr 2 mos

Average Tenure

1 yr 10 mos

Current Experience

Parallel web systems inc.

Member of Technical Staff

Jun 2024 – Present · 1 yr 10 mos · San Francisco Bay Area · On-site

Building something new and exciting.

Matic

Research Engineer

Apr 2023 – May 2024 · 1 yr 1 mo · Mountain View, California, United States · On-site

Designed and implemented new Visual SLAM algorithms at Matic - an early stage home robotics startup. Our product is featured in several tech magazines, notably our privacy preserving efficient SLAM system:
https://techcrunch.com/2023/11/02/mantics-robot-vacuum-maps-spaces-without-sending-data-to-the-cloud/

Visual SLAM algorithmsCamera-only localizationModel trainingTuningQuantizationCamera sensor experimentation+2

Waymo

Software Engineer (Machine Learning)

Jun 2022 – Sep 2022 · 3 mos · Mountain View, California, United States

Developed behavior prediction methods to anticipate and circumvent dangerous driving behaviors.
The methods developed as part of my internship are deployed on the Waymo fleet for safer autonomous driving

Behavior prediction modelsMachine LearningDriving behavior analysisArtificial Intelligence (AI)

Stanford university

Graduate Student Researcher

Oct 2021 – May 2022 · 7 mos · Stanford, California, United States

Implemented a novel machine learning based method for simulating millions of plasma particles efficiently.
Our method is 8x faster than classical particle solvers while being 6.8x more accurate than other deep learning methods.
Advisor: Prof Jure Leskovec
https://openreview.net/pdf?id=Rd68eTARk4

Machine learning methodsSimulation techniquesParticle solversMachine LearningComputer Science

Google

Research Associate

Jun 2021 – Aug 2021 · 2 mos · Bengaluru, Karnataka, India

Created a machine learning pipeline for personalized lifestyle interventions, with the goal of hypertension alleviation.
Incorporated interpretable time series models and counterfactual evaluation for better model interpretability.
Built the automated coaching engine for the real-world user experiment that demonstrates the efficacy of chatbot based fitness coaches as compared to human coaches. Won Best Paper Award at the conference.
https://dl.acm.org/doi/abs/10.1145/3503252.3531301

Machine learning pipelineTime series modelsCounterfactual evaluationMachine LearningArtificial Intelligence (AI)

Tower research capital

Quantitative Research Intern

Apr 2020 – May 2020 · 1 mo · Gurugram, Haryana, India

Designed and implemented machine learning & deep learning strategies to maximize profit and improve precision for aggressive algorithmic trading.

Machine learning strategiesDeep learningAlgorithmic tradingMachine LearningAlgorithmic Trading

Technische universität braunschweig

Research Intern

May 2019 – Jul 2019 · 2 mos · Brunswick, Lower Saxony, Germany

Designed and implemented a compressed-sensing based algorithm to estimate Direction of Arrival (DoA) for Terahertz signals and reduce measurement acquisition time by 61%. These signals are used for 6G communications.
Advisor: Prof Thomas Kuerner
https://ieeexplore.ieee.org/abstract/document/9411406

Compressed-sensing algorithmsDirection of Arrival estimationSignal ProcessingMachine Learning