Software engineer building production AI and backend systems. I work across low-latency APIs, durable workflows, retrieval infrastructure, and data pipelines, with a biotechnology background that helps when software meets scientific data.
Based in Boston, Massachusetts.
A Go control plane for expensive workflows that validates run intent, evaluates deterministic policy, and binds reviewer decisions to exact immutable specifications before execution.
Go
PostgreSQL
Nextflow
nf-core/rnaseq
Docker
AWS ECSRDSALB
Terraform
PrometheusDurable Go workflow engine with idempotency and crash recovery
A distributed-systems project built around exactly-once step execution, retries, saga compensation, Redis Streams, and event-backed recovery after worker crashes.
GoPostgreSQL
Redis StreamsDockerPrometheusCobra
YAML
GitHub Actions
Render
A Python and Go MCP service for agent-facing scientific retrieval that preserves verbatim source payloads and hash-chains each retrieval to the preceding audit record.
PythonGo
FastMCP
SQLite
TursoDockerGitHub Actions
Hugging Face Spaces
PyPIAuditable agent workflows for scientific evidence
A local-first AI system that traces generated claims through a durable workflow, bounded scientist agents, deterministic verification, and a content-addressed evidence graph.
Python
FastAPI
Qwen 3
OllamaSQLite
DuckDBPostgreSQL
RDKit
SHAP
Evo 2
Boltz-2
NVIDIA API
React
TypeScript
3Dmol.jsDockerHugging Face SpacesReliable glucose forecasting from streaming sensor data
A fault-tolerant data and ML pipeline that validates continuous-glucose-monitor readings, trains a Temporal Fusion Transformer, and serves 30- and 60-minute forecasts through FastAPI and ONNX Runtime.
Python
PyTorch
PyTorch Lightning
Kafka
TimescaleDB
PySpark
AirflowFastAPI
ONNX RuntimeDocker
GrafanaHugging Face SpacesA reliability-aware trust layer between perturbation screens and experimental decisions
A Python benchmarking harness for single-cell perturbation models that validates datasets, builds unseen-perturbation splits, scores models against strong baselines, and grades every prediction by how reliable the underlying measurement is.
PythonPyTorch
GEARS
Scanpy
AnnData
Pertpy
pandas
Plotly
Streamlit3Dmol.jsDockerServerless data and AI systems
Chicago Education Advocacy Cooperative
Deployed an EventBridge-triggered AWS Lambda and FastAPI pipeline across 20+ sources, saving 2 hours 40 minutes per newsletter cycle and reducing repeated Llama 3 calls by 40%.
Scientific retrieval and model adaptation
Indian Institute of Technology Dharwad
Built LipidOS with a six-person research team, expanding usable lipid assignments from 99 to 699 and delivering BGE/SPECTER2 retrieval over 12,563 text chunks and 1,058 spectral records.
Backend and data infrastructure
Gutslane Precitech
Reduced p95 API latency from 8 seconds to 200 milliseconds and built Airflow pipelines processing 10,000+ daily records, reclaiming more than 8 analyst hours per week.
OmicsLogic Inc.
Srikar Yadhunandan · IEEE FNWF 2024 · Explainable AI for interpreting energy-efficiency behavior in Open RAN systems · DOI: 10.1109/FNWF63303.2024.11028856