/

Steven R. Ness, Ph.D.

s n e s s AT s n e s s . n e t

Victoria, British Columbia, Canada

Zax Institute

sness.net LinkedIn Google Scholar Medium Github sness Github sness23 Sourceforge ResearchGate OpenReview ACM ScholarGPS IEEE AMiner

Summary

An interdisciplinary scientific programmer and computational biologist with a Ph.D. in Computer Science and over 30 years driving high‑impact research, product development, and AI‑powered biotech ventures. I combine deep domain expertise in structural biology and protein‑ligand docking with hands‑on machine‑learning and software‑engineering leadership.

Key Qualifications & Achievements

Founding AI‑Driven Biotech Platforms:

– Built and scaled folding.ai (2021–2025) and doi.bio (2025–Present), end‑to‑end intelligence‑augmentation systems that integrate generative models with GPU‑accelerated molecular viewers to triage and design novel therapeutics.

Enterprise ML & Infrastructure Leadership:

– As Lead Member of Technical Staff at Salesforce, designed and delivered production‑scale service‑mesh infrastructure (Envoy/Istio), voice‑AI assistants, and master‑data orchestration (Data Cloud) now serving millions of profiles.

Proven Scientific Programmer & Researcher:

– Authored CRANK (GPL‑licensed crystallography pipeline used by 100+ labs) and Gamma docking suite (University of Alberta), and published in Nature, Structure, and Biochemistry.

Cross‑Functional Collaboration & Wet‑Lab Integration:

– Partnered with protein designers and biologists throughout PhD (Orchive bioacoustic ML) and later roles to translate experimental feedback into model refinements, achieving high‑accuracy detection/classification and driving wet‑lab testing campaigns ([sness.net][1]).

Alignment with Latent Labs Responsibilities

Building ML Models for the Physical World :

I’ve rapidly prototyped generative and deep‑learning models—from orca‑call classification to AI‑driven protein design—and defined custom evaluation metrics tied directly to experimental success.

Data Curation & Metric Development:

The Orchive project involved curating 20 000+ hrs of audio to train ML pipelines and devising benchmark classification metrics; I’ve also managed large bioinformatics datasets and high‑throughput pipelines in genomics.

Collaborative Codebase & Infrastructure Maintenance:

I’ve led joint codebases with strict CI/CD, Docker/PM2 deployments, and cloud (AWS/GCP) architectures, ensuring robust, maintainable software across multi‑disciplinary teams.

Wet‑Lab Planning & Model Inference:

With hands‑on roles in structural biology labs (UBC, Leiden) and biotech startups, you’ve planned wet‑lab assays, performed in‑silico inference, and iteratively improved models based on real‑world feedback.

Continuous Learning & Knowledge Sharing:

I regularly publish at leading conferences (ICML, ISMIR), organize reading groups, and mentor across academia and industry, staying at the forefront of ML and cell‑biology advances.

💼 Professional Experience

Founder – doi.bio (Self‑employed)

Mar 2025 – Present (5 mo) · Victoria, BC (Hybrid)

Founder of doi.bio.

Founder & CEO – folding.ai

May 2021 – Mar 2025 (3 yr 11 mo) · Victoria, BC

Software Engineer – Quip

May 2020 – Apr 2021 (1 yr)

Lead Member Of Technical Staff, Einstein Voice Assistant – Salesforce Research

Nov 2019 – May 2020 (7 mo) · San Francisco, CA

Member of the Einstein Voice Assistant team (formerly MetaMind), building an AI‑based voice assistant smart‑speaker integrated with Salesforce core.

Lead Member Of Technical Staff, Service Mesh – Salesforce

Nov 2018 – Nov 2019 (1 yr 1 mo) · San Francisco, CA

Delivered Envoy proxy implementation and collaborated with Istio to deploy service‑mesh infrastructure for Salesforce Hyperforce / Customer 360.

Lead Member Of Technical Staff, Team Lead – Customer 360 – Salesforce

Feb 2016 – Nov 2018 (2 yr 10 mo) · San Francisco, CA

Founding member and UI Team Lead of the Customer 360 tiger team; presented demos to Salesforce founders and EVPs; partnered with architects to deliver the Customer 360 Data Manager product.

Senior Software Developer – Priologic Software Inc.

Oct 2013 – Oct 2014 (1 yr 1 mo) · Victoria, BC

Developed EasyRTC (open‑source WebRTC server/client SDKs) for web, iOS, and Android (GitHub).

Ph.D. Student

University of Victoria (Jan 2010 – Dec 2013 | 4 yrs) Victoria, BC

Projects:

Orchive – Data mining a massive bioacoustic archive of 20 000+ hrs of orca recordings from OrcaLab (1980 – present), with a web-based interface for researchers.

Cantillion – Designed and implemented the Cantillion interface (uploaded Oct 16, 2013).

Ocean Networks Canada (VENUS/NEPTUNE) – Sensor data integration and analysis.

M.Sc. Student

University of Victoria (2007 – Dec 2009 | 3 yrs) Victoria, BC

Projects:

Orchive – Early prototyping of the bioacoustic archive interface.

Cantillion – Initial interface design and implementation.

Audioscapes – Exploratory work on large-scale music analysis.

Software Engineering Intern

Google (News Performance Team)

May 2012 – Aug 2012 (4 mos)

Software Research Intern

Google Research (Human Auditory Models)

May 2011 – Aug 2011 (4 mos)

May 2010 – Aug 2010 (4 mos)

Scientific Programmer

Leiden University

Sep 2003 – Jan 2006 (2 yrs 5 mos)

Senior Programmer

BC Genome Sciences Centre

Feb 2000 – Sep 2001 (1 yr 8 mos)

Programmer

Stormix Technologies Inc.

Sep 1999 – Jan 2000 (5 mos)

Structural Biochemist Researcher & Software Engineer / Systems Administrator

University of British Columbia

May 1997 – Sep 1999 (2 yrs 5 mos)

Software Developer in Structural Biochemistry

University of Alberta

Jan 1995 – Apr 1997 (2 yrs 4 mos)

Structural Biochemistry Software Developer Intern

University of Alberta

Sep 1992 – May 1995 (2 yrs 9 mos)

Programmer Intern

Alberta Environment Research Centre

Jul 1987 – Aug 1987 (2 mos)

Education

University of Victoria

2010 – 2013 · GPA: 9.0

University of Victoria

2007 – 2010 · GPA: 9.0

Activities: Computer Science Course Union (community teaching), Society for Indian Classical Music

Volunteering

President — Society for Indian Classical Arts, University of Victoria

Jan 2010 – Dec 2013 (4 yr) · Arts & Culture

Lead Developer & Scientist — The Orchive

May 2007 – Jan 2025 (17 yr 9 mo) · Animal Welfare

Developed and maintained the Orchive website (23 000+ hours of orca vocalizations; bioacoustic archive annotation tools).

CORE COMPETENCIES

Publications

Full list: Google Scholar.

Integrated entity view across distributed systems

US Patent 10,936,582

ORCA-SLANG: An Automatic Multi-Stage Semi-Supervised Deep Learning Framework for Large-Scale Killer Whale Call Type Identification.

C Bergler, M Schmitt, AK Maier, H Symonds, P Spong, SR Ness, …

Interspeech, 2396-2400

User interface for commerce architecture

D Woodward, A Chadda, D Hacker, N Steven, M Lagrotte, J Moody, C Bill, …

US Patent App. 16/264,391

Orchive

S Ness

Human and machine annotation in the Orchive, a large scale bioacoustic archive

S Ness, G Tzanetakis

2014 IEEE Global Conference on Signal and Information Processing (GlobalSIP …

Taking a walk on the wild side: teaching cloud computing on distributed research testbeds

Y Zhuang, C Matthews, S Tredger, S Ness, J Short-Gershman, L Ji, …

Proceedings of the 45th ACM technical symposium on Computer science …

The Orchive: Data mining a massive bioacoustic archive

S Ness, H Symonds, P Spong, G Tzanetakis

arXiv preprint arXiv:1307.0589

The Orchive: A system for semi-automatic annotation and analysis of a large collection of bioacoustic recordings

S Ness

University of Victoria (Canada)

Sonophenology: A multimodal tangible interface for the sonification of phenological data at multiple time-scales

S Ness, P Reimer, J Love, WA Schloss, G Tzanetakis

Journal on Multimodal User Interfaces 5 (3), 123-129

On Computational Transcription and Analysis of Candences in Oral and Semi-Oral Chant Traditions

D Biro, P Van Kranenburg, S Ness, G Tzanetakis, A Volk

Proceedings of the Second International Conference on Analytical Approaches …

On computational transcription and analysis of oral and semi-oral chant traditions

DP Biró, P Van Kranenburg, SR Ness, G Tzanetakis, A Volk

Proceedings of the Analytical Approaches to World Music Conference, AAWM 2012

Stability and variation in cadence formulas in oral and semi-oral chant traditions–a computational approach

DP Biró, P van Kranenburg, SR Ness, G Tzanetakis, A Volk

Proceedings of the 12th International Conference on Music Perception and …

Auditory sparse coding

SR Ness, T Walters, RF Lyon

Music Data Mining. Boca Raton, FL, 33487-2742

Strategies for orca call retrieval to support collaborative annotation of a large archive

SR Ness, A Lerch, G Tzanetakis

2011 IEEE 13th International Workshop on Multimedia Signal Processing, 1-6

Music Information Robotics: Coping Strategies for Musically Challenged Robots.

SR Ness, S Trail, PF Driessen, WA Schloss, G Tzanetakis

ISMIR, 567-572

Automatic event detection for long-term monitoring of hydrophone data

F Sattar, PF Driessen, G Tzanetakis, SR Ness, WH Page

Proceedings of 2011 IEEE Pacific Rim Conference on Communications, Computers …

Controlling Real time sound spatialization using the radiodrum

SR Ness, G Odowichuk, PF Driessen, G Tzanetakis

ICMC 2011

A computational investigation of melodic contour stability in Jewish Torah Trope Performance Traditions

P Van Kranenburg, DP Biró, SR Ness, G Tzanetakis

Proceedings of the International Society on Music Information Retrieval …

Gesture Analysis of Radiodrum Data

SR Ness, S Methabi, G Odowichuk, G Tzanetakis, WA Schloss

ICMC 2011

Computer-assisted cantillation and chant research using content-aware web visualization tools

SR Ness, DP Biró, G Tzanetakis

Multimedia Tools and Applications 48 (1), 207-224

Sonophenology: a tangible interface for sonification of geo-spatial phenological data at multiple time-scales

SR Ness, P Reimer, N Krell, G Odowichuck, WA Schloss, G Tzanetakis

Geoshuffle: Location-Aware, Content-based Music Browsing Using Self-organizing Tag Clouds.

S Miller, P Reimer, SR Ness, G Tzanetakis

ISMIR, 237-242

Improving automatic music tag annotation using stacked generalization of probabilistic svm outputs

SR Ness, A Theocharis, G Tzanetakis, LG Martins

Proceedings of the 17th ACM international conference on Multimedia, 705-708

Assistive music browsing using self-organizing maps

G Tzanetakis, MS Benning, SR Ness, D Minifie, N Livingston

Proceedings of the 2nd International Conference on PErvasive Technologies …

Content-aware web browsing and visualization tools for cantillation and chant research

SR Ness, G Tzanetakis, DP Biró

2009 Seventh International Workshop on Content-Based Multimedia Indexing, 81-86

Assistive music browsing using self-organizing maps

S Ness

Proceedings of the 2nd International Conference on PErvsive Technologies …

New tools for use in the musicology of record production

K McNally, G Tzanetakis, SR Ness

Content-aware visualizations of audio data in diverse contexts

S Ness

MIREX 2010

S Ness, G Tzanetakis

### Somba: multiuser music creation using self-organizing maps and motion tracking

SR Ness, G Tzanetakis

ICMC

Audioscapes: exploring surface interfaces for music exploration

SR Ness, G Tzanetakis

ICMC

Chants and orcas: semi-automatic tools for audio annotation and analysis in niche dom

ains

SR Ness, M Wright, LG Martins, G Tzanetakis

Proceedings of the 2nd ACM workshop on Multimedia semantics, 9-16

Decoding the song: Histogram-based paradigmatic and syntagmatic analysis of melodic formulae in Hungarian laments, Torah trope, tenth century plainchant and Koran recitation

DP Biró, S Ness, WA Schloss, G Tzanetakis, M Wright

Phase refinement through density modification

JP Abrahams, JR Plaisier, S Ness, NS Pannu

Macromolecular Crystallography: conventional and high-throughput methods, 143

Extending the resolution and phase-quality limits in automated model building with iterative refinement

P Skubák, S Ness, NS Pannu

Biological Crystallography 61 (12), 1626-1635

Crank-New Methods in Automated Structure Solution

SR Ness, RAG de Graaff, JP Abrahams, NS Pannu

ACTA CRYSTALLOGRAPHICA A-FOUNDATION AND ADVANCES 61, C55-C55, 2005

CRANK: new methods for automated macromolecular crystal structure solution

SR Ness, RAG de Graaff, JP Abrahams, NS Pannu

Structure 12 (10), 1753-1761, 2005

CRANK-new methods for automated structure solution

S Ness, RAG de Graaff, JP Abrahams, NS Pannu

Foundations of Crystallography 60 (a1), s16-s16, 2004

A physical map of the mouse genome

SG Gregory, M Sekhon, J Schein, S Zhao, K Osoegawa, CE Scott, …

Nature 418 (6899), 743-750

Assembly of fingerprint contigs: parallelized FPC

SR Ness, W Terpstra, M Krzywinski, MA Marra, SJM Jones

Bioinformatics 18 (3), 484-485

Structural models for the alkaline conformers of yeast iso‐1‐ferricytochrome c

SR Ness, TP Lo, AG Mauk

Israel Journal of Chemistry 40 (1), 21-25

Structure-Based Design Guides the Improved Efficacy of Deacylation Transition State Analogue Inhibitors of TEM-1 β-Lactamase,

S Ness, R Martin, AM Kindler, M Paetzel, M Gold, SE Jensen, JB Jones, …

Biochemistry 39 (18), 5312-5321

Molecular Docking with a View: The Integration of a Monte Carlo Docking Program into a Virtual Reality Environment

TN Hart, RE Gillilan, R Lilien, SR Ness, RJ Read

High Performance Computing Systems and Applications, 309-322

Critical evaluation of the research docking program for the CASP2 challenge

TN Hart, SR Ness, RJ Read

Proteins: Structure, Function, and Bioinformatics 29 (S1), 205-209

GAMMA: a new docking program utilizing an advanced evolution system algorithm as an engine

S Ness, T Hart, R Read

Acta Crystallographica Section A 52, C90-C90, 1996

Monte Carlo algorithms for docking to proteins

RJ Read, TN Hart, MD Cummings, SR Ness

Supramolecular Chemistry 6 (1-2), 135-140


PATENTS


SELECT AWARDS & HONOURS


COMMUNITY & LEADERSHIP


LANGUAGES

English (native) - French (basic) • Dutch (basic) - Mandarin (basic)

Essay format

My name is Steven Ness and I have been doing ligand protein docking since before it was cool. In 1988 while in grade 9 for the science fair I developed Designer Genes, a program for Apple ][e and in later years Mac Plus that provided a number of bioinformatics tools for a given DNA sequence, including a device that I made to read electrophoresis sequencing gels that hooked up to the Apple ][e game port.

After success at the Canada Wide Science Fair, I was introduced to Dr. Michael N.G. James and Dr. Randy Read at the University of Alberta who sparked my love of structural biochemistry, I remember sitting in class and instead of focusing on class I was trying to get my eyes to defocus so I could see the wall-eyed stereo diagrams in x-ray crystallography papers, seeing a 3D molecule spring out of the 2D page was a revelation to me in grade 10.

After graduation, I went to the University of Alberta and during the summers I worked in the lab of Dr. Randy Read working on software to do protein-ligand docking and our team wrote the DockVision suite of programs that performed protein-ligand docking. I wrote a Genetic Algorithm docking method that was part of this software after conversations with Dr. John Moult and Colin Broughton.

We participated in the second ever CASP competition with our DockVision suite and some of our dockings were quite close to the crystal structure. The structures that we did the best on were the ones that we received the most feedback from crystallographers on, which convinced me that collaboration is essential in structural biology.

After graduation, I kept working for Dr. Randy Read and joined an exciting collaboration of many labs at the University of Alberta studying Hepatitis A that eventually led to a Nobel Prize in this field.

As part of this collaboration I had a number of roles including software developer, but one of the most impactful ones was that I was very well experienced with using the main software to display molecules including BioSym/Insight and Grasp from Anthony Nicholls, and I would drive as Dr. James, Dr. Malcolm and other scientists would give insights on what they saw.

This experience is what led me to create my new startup ci.doi.bio which is a collaboratative viewer for macromolecular structures.

I then followed one of the member of Dr. James lab, Dr. Natalie Strynadka who was starting up her own lab at the University of British Columbia. In her lab I worked on many projects involving x-ray crystallographic software, from building the open source software to developing new software. I also helped on a number of structures and determined structures and published these.

After this I worked at a Linux startup in Vancouver named Stormix during the dot-com boom and worked on low-level kernel code, developing one of the first graphical startup screens for a Linux distribution.

Subsequently I moved to the BC Genome Sequence center where I got to work with Dr. Michael Smith and his team on the Human and Mouse genome projects. I worked on the physical mapping software, speeding parts of it up by massive amounts, taking a O(N^2) algorithm that ran on program initialization and making it O(N), making software that took 3 hours to start on the full Human Genome dataset to take only seconds.

After this, I went to work with another student of Dr. Read, Dr. Navraj Pannu at Leiden University in the Netherlands. Here I developed CRANK, a software package that allowed a user to create a custom pipeline to automatially solve crystal structures. This software was incorporated into the CCP4 software suite which is installed in (almost) every macromolecular x-ray crystallography lab in the world.

I then moved back to Canada and started my Ph.D/M.Sc. i Computer Science at the University of Victoria working on applying Machine Learning techniques to music. My primary project was The Orchive, one of the largest collections of Orca vocalizations with over 30,000 hours of recorded sound. I developed Machine Learning systems to both detect and classify different Orca calls and achived 97% detection and 80% classification accuracy. For this I used many different techniques, including Deep Learning.

During this Ph.D. I was lucky to get three summers of internships at Google, the first two in Google Research where I worked on developing Machine Learning algorithms for cover song detection from audio features. The third was in Google News where I worked on measuring and optimizing latency in Google News.

After graduation, I worked at Priologic, a software startup in Victoria, BC. where I worked on EasyRTC, a framework to allow developers to develop WebRTC video conferencing applications.

I was then lucky to find a position at Salesforce on a new team directly under the then-CTO of Salesforce, Walter Macklem on a bold new plan to break up the monolith of Saleforce into a microservices architecture.

This led me to work on the most exciting project I have been on, a Tiger Team composed of some of the best people from all of the companies Salesforce had acquired, making the first demo of the product that now powers all of the integrations between all of the sub-companies at Salesforce. This then led me to work on the Istio team and to work closely with the Kubernetes team, part of this transformation of Salesforce.

I then moved to the Salesforce Einstein team where I got to work on the Einstein Smart Speaker, an early app that was similar to ChatGPT, but a few years early.

This entire team then moved into the Quip team, a startup acquired by Salesforce that made an HTML editor similar to Notion. In this team I developed software to help integrate this new startup with the Salesforce monolith.

Alphafold2 had just been released, and one month after this we started folding.ai, a startup that worked on developing software in this field. We had a number of products, the most exciting of which was one that used one of the attention heads in the "BERTology Meets Biology: Interpreting Attention in Protein Language Models" model to predict protein-ligand binding sites.

After this startup folded, I started doi.bio, a website that allows AIs to read papers and information in structural biochemisty to help drive protein-ligand docking.