working on a protocol to co-ordinate distributed, and autonomous devices(such as robots and drones) using a leader-follower paradigm. Implementing Raft to handle consensus leader election, takeover and tiebreakers. Using TLA+ to perform model checking and validate the formal model of the protocol. Learn more: https://github.com/dhilanpatel26/leader_follower

Investing upto $250K in early-stage student-founded startups, powered by General Catalyst.If you are a technical student founder, especially if you are building in data/ml/ infra + devtools, shoot me a note would love to chat -> bnjoroge@roughdraft.vc

investing in data/ml infra, devtools, and security.

$3M AUM student-run and led VC firm that invests in early stage, capital-efficient companies with ties to Johns Hopkins.Unicorns:-Elemy, $219M Series B placing valuation at $1.15B (https://www.elemy.com/)Portfolio Exits:-KITT.AI, acquired by Baidu (http://kitt.ai/)-Birdi, acquired by Luma (http://birdihome.com/)-Tueo Health, acquired by Apple-Progressly, acquired by Box (https://progressly.com)-WiseBanyan, acquired by Axos Financial (https://wisebanyan.com) - alevelcapital.com/portfolioHopkins-affiliated founders/early employees of early stage startups, please share your deck at bill{at}alevelcapital{dot}com

dev platform team. wrote k8s controllers using Go to automate the registration of new products and services using a gitops-native approach. Also contributed to idpbuilder, a self-service internal dev platform used by AWS, Adobe, Salesforce, etc. Go, K8s, Kustomize, Kubebuilder, ArgoCD etc

productionized AI models for first-party msft applications.PyTorch + FastAPI + Azure + ONNX + abit of C#/dotnet.

crypto infra sourcing/diligence + building internal eng tools for platform/ops team. Diligence:pyth.network(on-chain real-time market data solution)Contributing to https://twosigmaventures.com/open-source-index/ for crypto projects.

ml infra + devtools. trained and deployed(across multiple GPU nodes) an end-to-end rcnn neural network using pytorch and opencv that did object recognition and instance segmention on gigabytes of synthetic vision data, demonstrating the validity of the project.built middleware that enabled the team to programmatically launch workflows with much better composability, less dependency and better integration(homegrown orchestration tooling)used Dask and PySpark to build pipelines that parallelized multiple workflows(such as synthetic data generation and simulation), reducing total compile time by 20% and supporting 8x more data sizes. reproduced multiple research studies, mostly on embodied ai and computer vision, and presented to the team. python, dask, pyspark, docker etc.

Founding dev for a on/off ramp rails for crypto/fiat transfers. Built on Celo Blockchain, and wrote v1s of the API and sdks.

Developed a custom NodeJS app that interfaced Sokowatch’s data systems, Botpress Framework and an external Whatsapp Client. A/B tested with 200 users following Agile SDLC, increasing CSAT by 2 points.• Built a light-weight beta version of a PWA using Flask that allowed users to submit, and track their orders with support for background async/fetch/push notifications etc. A/B tested with 200 users and received a NPS score of 8.2• Wrote and validated user stories for Sokowatch’s Referral system, reducing churn by 5% .

Worked with a team of devs and PM to build an end-to-end, dockerized, and version-controlled prototype of a Client and Donor Relationship Management platform running multiple worker processes, and exposed as a REST API consumed by an interactive Frontend Dashboard built with chart.js. Shipped to 800+ users.Wrote test suite, and custom Python scripts for Quantibly's ERP'S 3rd Party API integrations.Tracked and evaluated the platform's metrics(CAC, LTV, CPC, Churn rate, Stickiness Ratio) using Google Analytics to prepare an evaluatory report on the need for increased R&D into the feature.Tech stack: Python, Flask(and additional extensions), MySQL, ElasticSearch, Redis Cache, Docker, Nginx, EC2 instance, S3, RDS, and RQ for the background jobs.