Resume
Tessa Everett
MIT AI researcher focused on ML for protein design, structure prediction, and data-driven drug discovery.
EDUCATION
Massachusetts Institute of Technology (Jan 2026)
Candidate for a B.S. in Artificial Intelligence and Decision-Making
Computing Coursework: Machine Learning, Computer Vision, Inference & Probability, Optimization Methods, Design and Analysis of Algorithms, Data Science, Linear Algebra
Biology Coursework: Generative AI in Biology, AI in Medicine, Biochemistry, Organic Chemistry, Thermodynamics of Biomolecular Systems
WORK EXPERIENCE
Undergraduate Researcher – AI for Protein Design (August 2025–Present)
Keating Lab, MIT, Cambridge, MA
- Developing AI-driven pipelines to design orally bioavailable mini-proteins for therapeutic targets.
- Building quantitative conformational scoring metrics (RMSD distributions, helical content, ensemble compactness) to evaluate peptide stability and design candidates.
- Leveraging MD/BioEmu to model unbound conformational ensembles and extract stability and flexibility signatures.
Computer Vision and Data Science Intern (May 2025–August 2025)
Syght, Boulder, CO
- Built a reproducible, serverless training pipeline on Google Cloud (Vertex AI) with automated hyperparameter tuning for faster iteration.
- Trained real-time threat detection models, fine-tuning pretrained RGB CNNs on mm-wave images.
- Doubled dataset size. Designed data collection & augmentation to maximize performance while balancing resources.
ML Engineering Intern (May 2024–September 2024)
SeeScan, San Diego, CA
- Built a dataset pipeline with automated class rebalancing, improving performance on underrepresented classes.
- Migrated training workflows from local machines to AWS, upgrading the pipeline for improved accuracy and speed.
- Optimized computer vision models for small-object detection through systematic hyperparameter tuning.
Protein Engineering Intern (June 2023–July 2023)
Centre for Biotechnology and Bioengineering, Santiago, Chile
- Purified target proteins using liquid chromatography and conducted Bradford assays to assess protein functionality.
- Monitored and optimized batch-fed Pichia pastoris yeast cultivation for improved protein production.
TECHNICAL PROJECTS & COMPETITIONS
Protein Function Prediction - CAFA Challenge
- Built an end-to-end pipeline using ESM-2 and T5 embeddings with taxonomic features to predict GO terms.
- Trained a non-negative PU MLP classifier with homology-based splits for strong generalization and 5-fold ensembling.
- Implemented evaluation using IC-weighted F1 scores and hierarchical GO term propagation.
Inference-Time Scaling for 3D Diffusion Models
- Proposed and implemented a dynamic inference-time scaling algorithm combining adaptive Best-of-N sampling and a multimodal LLM verifier to enhance 3D diffusion model performance.
- Designed and executed an experiment on the Objaverse-XL dataset, demonstrating statistically significant improvements across perceptual fidelity metrics (PSNR ↑1.647, LPIPS ↓0.0855, SSIM ↑0.0283, FS ↑0.0187).
- Built custom evaluation pipelines using ICP alignment and 2D view synthesis; wrote statistical testing suite to validate gains.
- Achieved visual and numerical fidelity gains rivaling recent architectural improvements, with zero retraining overhead.
Spatial Transcriptomics Disease Prediction Challenge
- Developed a deep learning pipeline (ResNet50 + MLP) for disease classification using spatial transcriptomics images and gene expression data
- Engineered efficient data processing workflows with Dask, SpatialData, and anndata, enabling scalable preprocessing
- Ranked 30th globally in a competitive biomedical ML challenge hosted by CrunchLabs.
SKILLS
Languages & Tools: Python (NumPy, Pandas, PyTorch, Scikit-learn), Julia, Linux Systems
Machine Learning: diffusion models, CNNs/MLPs, transformers, classical ML, optimization, statistical modeling
Biology & Bioinformatics: AlphaFold/Boltz, RFdiffusion, ESM-2, UniProt/GO, MDAnalysis, BioPython, BioEmu
Cloud Tools & Infrastructure: Google Cloud, AWS, Docker, SQL basics, Git/GitHub
EXTRACURRICULARS
Alpha Delta Phi Literature Society (September 2022–Present)
- Executive board member, managing up to 50+ residents and $400,000 annual revenue.
Varsity Track and Cross Country Athlete (August 2022–August 2025)
- Dedicated member of MIT's Varsity Track and Cross Country teams, earning NEWMAC athlete of the week in 2022.
MIT Strategic Game Society (January 2024–Present)
- Regularly engaging in competitive strategy games, refining probabilistic and adaptive thinking.