Data-to-Insight for R&D
Turning sparse, noisy, high-dimensional experimental data into actionable models - especially oligonucleotide, analytical, and multi-modal lab datasets.
Charles Phillips, PhD
Principal Solution Scientist
Scientific AI & ML · Multimodal Data · Computational Chemistry
I build practical ML systems that turn complex experimental and scientific data into working solutions - combining modelling, software engineering, and clear communication for R&D teams in life sciences, chemistry, and materials.
Track Record
2+
Years in Scientific AI
Intellegens · May 2024–present
80+
Live Demos & Trainings
Global scientific audiences
1
Journal Covers
Peer-reviewed · 2026
8+
Years of Coding
Since 2018
Background
I'm a scientist and ML practitioner focused on measurable impact: faster process development, better use of messy experimental data, and systems that research teams can actually adopt. My path runs from an MChem at Strathclyde and polymer research at Huntsman (granted patent) through a Cambridge MPhil and PhD in computational chromatin biophysics, to industry delivery at Intellegens.
At Intellegens I led development of the Alchemite™ Oligonucleotide platform - from scientific strategy and implementation through product leadership to my current role as Principal Solution Scientist, designing multi-modal pipelines and bespoke solutions for enterprise R&D customers.
That work contributed to Intellegens being named a finalist for Life Science Company of the Year at the 2026 Cambridge Independent Science & Technology Awards. My PhD modelling underpins the Science Advances front-cover paper on linker histone H1 and chromatin organisation.
I use whatever approach fits the problem - coarse-grained simulation, classical ML, platform engineering, or careful data pipeline design - and I care about explaining results clearly to scientists, engineers, and commercial stakeholders.
Areas of Expertise
ML Systems & Pipelines
Building extensions to Alchemite™, bespoke preprocessing, automation frameworks, and integrations across vendor platforms and custom infrastructure.
Computational Biophysics
Multiscale chromatin modelling, phase behaviour, enhanced sampling (LAMMPS), and linking molecular simulation to experimental biology.
Enterprise Scientific Delivery
Customer trials and pilots, data-quality mitigation, live technical demos, and PoC-to-product collaboration with development and commercial teams.
Technical Demos & Training
Translating complex ML architectures into insights aligned with research objectives - for pharma, biotech, and materials audiences.
Thought Leadership
Conference talks, panels, posters, and technical writing on ML for oligonucleotides, process development, and scientific software.
Recent Work
Selected outputs - from peer-reviewed research, ChemRxiv preprints, and Intellegens product work. Full detail on my CV.
Research · Science Advances, 2026
Linker histone H1 as liquid-like chromatin glue
Role: Computational modelling (PhD work); multiscale chromatin simulations supporting live-cell mechanistic framework.
Impact: Front-cover publication; bridges years of simulation development with experimental collaboration.
Read write-up · PaperProduct · Intellegens, 2024–2026
Alchemite™ Oligonucleotide Manufacturing
Role: Lead scientific/technical developer → innovation lead; predictive ML for synthesis, purification, and analytics.
Impact: Adoption by major industry partners; awards finalist; technical paper and recorded webinar.
Product · Technical paperPreprint · ChemRxiv, 2025
The Potential of Machine Learning in Oligonucleotide Therapeutics Manufacturing
Role: Co-author (Intellegens); ML for SPOS process development, impurity analysis, and manufacturing optimisation.
Impact: Open-access preprint from the Intellegens–CPI oligonucleotide collaboration - summarises how machine learning can accelerate therapeutic oligo manufacturing.
ChemRxiv · DOIFrom Problem to Production
I combine scientific framing, modelling, and engineering to deliver solutions that work in real R&D environments.
Frame the Science
Clarify the experimental question, data constraints, and what success looks like for the team.
Build Proof of Value
Prototype pipelines and models quickly - pilots, demos, and focused analyses before scaling.
Handle Messy Data
Automated preprocessing, quality checks, and validation - especially where modalities and vendors differ.
Deliver & Hand Over
Integrate into customer workflows, train users, and collaborate on productisation where impact is clear.
Technology Stack
Tools and domains I work with regularly. Expand each category for detail.
Machine Learning & Data
Python (NumPy, Pandas, scikit-learn, PyTorch) · Alchemite™ · Gaussian processes & ensembles · Feature engineering · Image analysis · Leakage-aware validation · Multi-modal lab pipelines
Simulation & Computational Chemistry
LAMMPS · Coarse-grained MD · Enhanced sampling · Chromatin phase behaviour · C++ scientific computing · Free-energy methods · Oligonucleotide & polymer chemistry
Engineering & Delivery
Git · Docker · Bespoke automation · Customer trials & pilots · Technical demos · Cross-functional product collaboration
Domains
Oligonucleotide therapeutics · Biopharmaceutical manufacturing · Chromatin & epigenetics · Materials & chemicals R&D · Analytical & process data
FAQ
What industries have you worked in?
Life sciences and oligonucleotide manufacturing (primary), plus materials science, chemicals, and broader manufacturing through Intellegens customer work and my polymer research background.
Do you only do machine learning?
No. I work across ML, scientific software, molecular simulation, and solution design - choosing the tool that matches the data and the decision teams need to make.
What's your technical background?
MChem (Strathclyde), MPhil Scientific Computing (Cambridge), PhD in theoretical/computational chemistry (Cambridge, Collepardo group), and industry roles at Intellegens from ML Scientist through to Principal Solution Scientist.
What makes your approach different?
Depth in both simulation and ML, plus hands-on enterprise delivery - I understand experimental workflows, not just models, and I focus on adoption through demos, training, and robust data pipelines.
Beyond Work
Rowing - Won a rowing blade for Queens' College, Cambridge in May Bumps (M1, first division), beating St John's, Robinson, Jesus, and Trinity Cambridge colleges
Music - Actively play in bands; competed in Young Musician of the Year (2013 & 2014) for classical guitar; play seven instruments
Hobbies - MMA, climbing, wild swimming, and hiking
Get in Touch
Questions about my work or collaboration? Reach out via email or LinkedIn.