The default language of modern AI engineering. Type hints, async, strong stdlib.
DataMLAIAgentsMLOpsSafety
Tools catalog
Your full AI engineering stack.
81+ production tools — every layer of the stack, taught across the six schools. Filter by school, category, or how you use the tool.
81 of 81
Catalog
Every tool Safua learners touch.
Filters81 tools
By school
By category
By depth
Typed JavaScript for AI-app front ends and Node services that call AI APIs.
AIAgentsSystems-grade language for latency-sensitive inference and control planes.
MLOpsConcise, concurrent language powering a large share of cloud-native infrastructure.
MLOpsAIThe lingua franca of data. Windowing, joins, query plans — well beyond SELECT *.
DataAIMLThe reliable default for transactional data. Strong ecosystem, battle-tested.
DataAIMLOpsIn-memory key-value + caching. The reflexive choice for hot paths and queues.
DataAIMLOpsClassic relational engine still running a sizeable fraction of the internet.
DataDocument database. Used where schemas are fluid and nested documents are the unit.
AIEmbedded SQL engine. Perfect for local development and eval harnesses.
DataWide-column distributed database for write-heavy, globally replicated workloads.
DataFull-text search + analytics engine. Supports hybrid text + vector retrieval.
DataAICloud data warehouse separating compute from storage for elastic analytics.
DataIn-process analytical database. Run warehouse-grade queries inside a notebook.
DataMLColumnar analytical database built for sub-second queries over huge tables.
DataS3-compatible object storage you can run on your own hardware.
MLOpsDataVector search inside Postgres. One database, embedding-aware retrieval.
AIAgentsPurpose-built vector database with filtering and payload storage.
AIAgentsVector database with built-in reranking and multi-tenant retrieval.
AIAgentsDeveloper-friendly embedding store that runs in-process or as a service.
AIAgentsDistributed vector database designed for billion-scale similarity search.
AIAuthor, schedule, and monitor batch data pipelines as code.
DataMLOpsDistributed log and streaming platform. The backbone of event-driven systems.
DataMLOpsDistributed compute engine for big-data batch and streaming workloads.
DataMLTransform data in the warehouse with versioned, tested SQL models.
DataPython-native workflow orchestration with strong retry + observability defaults.
DataMLOpsDurable execution for long-running workflows, agent state, and human-in-the-loop.
MLOpsAIDistributed stream-processing engine for stateful, low-latency streaming.
DataDeep-learning framework with dynamic graphs. The research default turned production.
MLAISafetyDeep-learning framework with a strong production-serving story.
MLClassical ML: linear models, trees, SVMs, calibration — the rigorous baseline.
MLDataTransformers library + model hub. Standard for modern NLP + multimodal.
MLAIAgentsDataFrames for data manipulation and exploration in Python.
DataMLN-dimensional arrays and the math that every ML library depends on.
MLDataGradient boosting library. Still the thing that wins most tabular competitions.
MLDataFast gradient-boosting framework tuned for large feature spaces.
MLComposable transformations of numerical code. Auto-diff + XLA for scale.
MLHigh-level wrapper around PyTorch that removes training-loop boilerplate.
MLHyperparameter optimisation framework with pruning and parallel trials.
MLHigh-level deep-learning API, now unified across TensorFlow, PyTorch, and JAX.
MLNotebook environment for exploration, teaching, and reproducible reports.
MLDataSafetyTurn a Python script into a demo UI in minutes. Great for internal tools.
MLAIExperiment tracking, model registry, and deployment tooling — open-source.
MLMLOpsExperiment tracking + model registry with collaborative dashboards.
MLMLOpsTraining-run visualisation for losses, gradients, embeddings, and images.
MLContainer runtime for reproducible application packaging and local dev.
MLOpsAIDataContainer orchestration for scheduling, scaling, and networking services.
MLOpsCI/CD for test, build, and deploy workflows directly in your repo.
MLOpsInfrastructure-as-code for provisioning cloud resources reproducibly.
MLOpsPackage manager for Kubernetes: parameterised, versioned app deployments.
MLOpsGitOps continuous delivery for Kubernetes — the manifest in the repo is the truth.
MLOpsDashboards + alerting over metrics, logs, and traces. The visualisation layer.
MLOpsDataTime-series metrics database with pull-based collection and PromQL.
MLOpsUnified observability platform — metrics, traces, logs, RUM, and more.
MLOpsError tracking + performance monitoring for applications and LLM features.
MLOpsAIVendor-neutral standard for traces, metrics, and logs across stacks.
MLOpsDistributed tracing system for following requests across services.
MLOpsFramework for LLM applications: chains, tools, retrieval, agents.
AIAgentsData framework for LLM apps — ingestion, indexing, retrieval, query engines.
AIAgentsOfficial client library for the OpenAI HTTP API — installs as a package.
AIAgentsOfficial client library for the Anthropic HTTP API — installs as a package.
AIAgentsSafetyRun open-weight LLMs locally with one binary. Great for air-gapped evaluation.
AISafetyMulti-agent orchestration framework with conversational and tool-use primitives.
AgentsRole-based multi-agent framework with explicit tasks, tools, and memory.
AgentsGraph-based agent orchestration with typed state and durable execution.
AgentsTrace + evaluate LLM-application runs with structured datasets and graders.
AIAgentsStructured output validation and safety checks for LLM applications.
SafetyAIShapley-value feature attribution for any model. The interpretability default.
SafetyMLLocal interpretable model-agnostic explanations for individual predictions.
SafetyDeclarative data-quality tests that travel with your pipelines.
DataSafetyHTML sanitisation for LLM output bound for user-facing surfaces.
SafetyRuntime data validation via type annotations. The backbone of safe JSON APIs.
AISafetyDataPython test framework with fixtures, parametrisation, and plugin ecosystem.
DataMLAIMLOpsSafetyOpen-source vulnerability scanner for containers, filesystems, and IaC.
MLOpsSafetyService mesh for traffic management, security, and observability in Kubernetes.
MLOpsHigh-performance edge and service proxy — the data plane behind many meshes.
MLOpsSecrets management, encryption-as-a-service, and identity-based access.
MLOpsSafetyService discovery, service mesh, and config management for distributed systems.
MLOpsWeb server, reverse proxy, and load balancer that quietly runs the internet.
MLOpsPython web framework with type-driven validation. Great for AI-app backends.
AIMLOpsJavaScript runtime for backends, serverless functions, and LLM-app BFFs.
AI
Don’t see your tool? Custom missions can incorporate additional tools through the platform’s extensibility system. Talk to us.
Popular by school
What each school ships with.
Popular in Data Engineering
The default language of modern AI engineering. Type hints, async, strong stdlib.
DataMLAIAgentsMLOpsSafetyThe lingua franca of data. Windowing, joins, query plans — well beyond SELECT *.
DataAIMLThe reliable default for transactional data. Strong ecosystem, battle-tested.
DataAIMLOpsAuthor, schedule, and monitor batch data pipelines as code.
DataMLOpsDistributed log and streaming platform. The backbone of event-driven systems.
DataMLOpsContainer runtime for reproducible application packaging and local dev.
MLOpsAIDataDashboards + alerting over metrics, logs, and traces. The visualisation layer.
MLOpsData
Popular in Machine Learning
The default language of modern AI engineering. Type hints, async, strong stdlib.
DataMLAIAgentsMLOpsSafetyThe lingua franca of data. Windowing, joins, query plans — well beyond SELECT *.
DataAIMLDeep-learning framework with dynamic graphs. The research default turned production.
MLAISafetyTransformers library + model hub. Standard for modern NLP + multimodal.
MLAIAgentsExperiment tracking, model registry, and deployment tooling — open-source.
MLMLOps
Popular in AI Engineering
The default language of modern AI engineering. Type hints, async, strong stdlib.
DataMLAIAgentsMLOpsSafetyTyped JavaScript for AI-app front ends and Node services that call AI APIs.
AIAgentsThe lingua franca of data. Windowing, joins, query plans — well beyond SELECT *.
DataAIMLThe reliable default for transactional data. Strong ecosystem, battle-tested.
DataAIMLOpsPurpose-built vector database with filtering and payload storage.
AIAgentsDeep-learning framework with dynamic graphs. The research default turned production.
MLAISafetyTransformers library + model hub. Standard for modern NLP + multimodal.
MLAIAgents
Popular in Agentic AI
The default language of modern AI engineering. Type hints, async, strong stdlib.
DataMLAIAgentsMLOpsSafetyTyped JavaScript for AI-app front ends and Node services that call AI APIs.
AIAgentsPurpose-built vector database with filtering and payload storage.
AIAgentsTransformers library + model hub. Standard for modern NLP + multimodal.
MLAIAgentsFramework for LLM applications: chains, tools, retrieval, agents.
AIAgentsGraph-based agent orchestration with typed state and durable execution.
Agents
Popular in MLOps & Infrastructure
The default language of modern AI engineering. Type hints, async, strong stdlib.
DataMLAIAgentsMLOpsSafetyThe reliable default for transactional data. Strong ecosystem, battle-tested.
DataAIMLOpsAuthor, schedule, and monitor batch data pipelines as code.
DataMLOpsDistributed log and streaming platform. The backbone of event-driven systems.
DataMLOpsExperiment tracking, model registry, and deployment tooling — open-source.
MLMLOpsContainer runtime for reproducible application packaging and local dev.
MLOpsAIDataContainer orchestration for scheduling, scaling, and networking services.
MLOps
Popular in AI Safety & Governance
The default language of modern AI engineering. Type hints, async, strong stdlib.
DataMLAIAgentsMLOpsSafetyDeep-learning framework with dynamic graphs. The research default turned production.
MLAISafety
Faculty
Tools alone don’t graduate an engineer.
Named reviewers grade how you use these tools across every school.
EA
Dr. Emeka Adeyemi
Dean
School of Data Engineering
“Data is the foundation. Everything else is built on top of what you build here.”
M
Marcus
Lead Instructor
School of Data Engineering
“Tell me WHY before you show me HOW.”
T
Tomás
Teaching Assistant — SQL & Databases
School of Data Engineering
“Before you write the query, draw the tables. What connects them?”
R
Rina
Teaching Assistant — Pipeline Architecture
School of Data Engineering
“What happens when this task fails at 3am? Show me your retry logic.”
K
Kwame
Teaching Assistant — Data Modeling
School of Data Engineering
“Model for the questions people will ask, not the data you have.”
SL
Dr. Sarah Lin
Dean
School of Machine Learning
“A model that works but can’t be explained is a liability, not an asset.”
P
Priya
Lead Instructor
School of Machine Learning
“Before you tune hyperparameters, tell me why you chose this architecture.”
D
David
Teaching Assistant — Classical ML
School of Machine Learning
“Try a simple baseline first. You’d be surprised how often it wins.”
M
Mei
Teaching Assistant — Deep Learning
School of Machine Learning
“Think of attention as the model asking: which parts of the input matter most for this output?”
A
Alejandro
Teaching Assistant — Computer Vision
School of Machine Learning
“Your model is accurate. Now make it run in 50ms on a device with 2GB RAM.”
AO
Dr. Amara Osei
Dean
School of AI Engineering
“AI engineering is not about the model. It’s about the system around the model.”
K
Kaia
Lead Instructor
School of AI Engineering
“But does it scale?”
N
Nadia
Teaching Assistant — RAG & Retrieval
School of AI Engineering
“If your retrieval is wrong, your generation is confidently wrong. Fix retrieval first.”
H
Hiroshi
Teaching Assistant — Fine-Tuning
School of AI Engineering
“Show me your training data before you show me your hyperparameters.”
Z
Zara
Teaching Assistant — AI App Architecture
School of AI Engineering
“What’s the cost per request? Your architecture is only viable if the unit economics work.”
RP
Dr. Rashid Patel
Dean
School of Agentic AI
“You’re not building a tool. You’re building something that makes decisions. Treat that seriously.”
S
Soren
Lead Instructor
School of Agentic AI
“What happens when agent A and agent B disagree? Design for conflict, not just cooperation.”
Y
Yuna
Teaching Assistant — Agent Frameworks
School of Agentic AI
“Build a tiny agent first. Make it work. Then make it smart.”
E
Eliot
Teaching Assistant — Memory & Planning
School of Agentic AI
“An agent without memory is just a function call. Memory is what makes it an agent.”
CW
Dr. Chen Wei
Dean
School of MLOps & Infrastructure
“If it’s not in production with monitoring, it doesn’t exist.”
V
Viktor
Lead Instructor
School of MLOps & Infrastructure
“Your model is only as reliable as your deployment pipeline. Show me the pipeline.”
F
Fatima
Teaching Assistant — Model Serving
School of MLOps & Infrastructure
“What’s your p99 latency? What’s your cost per prediction? Those two numbers define your architecture.”
A
Andrei
Teaching Assistant — Monitoring
School of MLOps & Infrastructure
“If your alert fires, you’re already late. Design monitoring that predicts the problem.”
AN
Dr. Adaeze Nwosu
Dean
School of AI Safety & Governance
“The question is never just "can we build it?" It’s "should we, and how do we build it responsibly?"”
J
James
Lead Instructor
School of AI Safety & Governance
“Compliance is not paperwork. It’s architecture. Build it into the system from day one.”
L
Lucia
Teaching Assistant — Explainability
School of AI Safety & Governance
“If a patient asks why the model flagged them, what do you say? That’s explainability.”
O
Omar
Teaching Assistant — Red-Teaming
School of AI Safety & Governance
“I found three ways to make your model produce harmful output. Now let’s fix all three.”
YT
Dr. Yuki Tanaka
CTO
NovaMind AI
“Would you deploy this to production with your name on it?”
DV
Dimitri Volkov
VP of Engineering
NovaMind AI
“This isn’t a coding exercise. The retrieval team is blocked until your pipeline ships.”
RM
Rafael Mendes
Senior Engineer — RAG & Retrieval
NovaMind AI
“Your chunking strategy tells me everything about how you think about this problem.”
IN
Isla Nakamura
Senior Engineer — Agent Orchestration
NovaMind AI
“An agent that can’t explain its own decision is an agent you can’t trust.”
AR
Amir Rezaei
Tech Lead
NovaMind AI
“What’s the smallest thing you can ship today that moves us forward?”
MO
Dr. Miriam Okafor
CTO
DataForge Labs
“Show me how this handles a million records. Then we’ll talk about your algorithm.”
MC
Marcus Chen
VP of Engineering
DataForge Labs
“If you can’t monitor it, you can’t ship it.”
JP
Jin Park
Senior Engineer — Pipelines
DataForge Labs
“I evaluate idempotency before business logic.”
CJ
Clara Johansson
Senior Engineer — MLOps
DataForge Labs
“A notebook is not a deployment. Show me the Docker file, the CI pipeline, and the rollback strategy.”
LM
Luis Morales
Tech Lead
DataForge Labs
“If your teammate can’t understand this code in 6 months, rewrite it.”
AM
Dr. Andreas Müller
CTO
VisionArc
“Fast enough is never fast enough. Find the bottleneck and eliminate it.”
PS
Priyanka Sharma
VP of Engineering
VisionArc
“You have 2GB of RAM and 100ms latency budget. Make it work within those constraints.”
FD
Dr. Fatou Diallo
Senior Engineer — Computer Vision
VisionArc
“Explain the receptive field of your architecture. If you can’t, you don’t understand your model.”
KY
Kai Yamamoto
Senior Engineer — Deep Learning
VisionArc
“The best architecture is the simplest one that meets your requirements. Start there.”
AP
Aisha Patel
Tech Lead
VisionArc
“What’s the smallest thing you can ship today?”
EO
Dr. Elizabeth Okafor
CTO
Sentient Health
“Behind every data point is a patient. Build systems worthy of that responsibility.”
DN
Daniel Nakamura
VP of Engineering
Sentient Health
“Every line of code in healthcare is auditable. Write it like a regulator is reading it.”
LK
Dr. Lena Kowalski
Senior Engineer — Compliance & Safety
Sentient Health
“In healthcare, a logging mistake isn’t a bug — it’s a lawsuit.”
RM
Dr. Ravi Mehta
Senior Engineer — Explainability
Sentient Health
“Your model denied a patient coverage. Can you explain why to their doctor? That’s the standard.”
ST
Sophie Tremblay
Tech Lead
Sentient Health
“We ship slower than other companies. But what we ship never harms a patient.”
JH
Jordan Hayes
Career Director
Safua Career Office
“Your profile says more than your resume ever could.”
One catalog. Every skill you need to ship.
Real tools, real workflows, real reviews from named faculty.