Agentic frameworks landscape tiles over a teal tech backdrop

Agentic frameworks available today

Survey of agentic frameworks with principal-level selection criteria: LangGraph, LlamaIndex, CrewAI, provider SDKs (OpenAI, Claude, Google ADK, Microsoft), Pydantic AI, Mastra, lock-in trade-offs, migration playbooks, and interview decision trees.

Frameworks vs architectures

The agent ecosystem exploded: open-source orchestrators, RAG toolkits, role-based "crews," and first-party SDKs from OpenAI, Anthropic, Google, and Microsoft. Naming ten libraries impresses nobody. Showing you know which control model each owns — and when to skip frameworks entirely — does. For a focused Q&A on Deep Agents vs LangChain vs LangGraph and the app types you can build, see Deep Agents, LangGraph & LangChain — Q&A.

Read this after Agentic architectures and Agentic patterns. Architecture chooses the loop; the framework implements it. Pair with RAG and Prompt engineering for the surrounding stack.

Six buckets of agent frameworks by job: workflows, RAG, crews, provider SDKs, typed/enterprise, TypeScript.
Organize by job-to-be-done, not by which blog ranked them "best."
01Graphs

LangGraph control plane.

02Data

LlamaIndex · Haystack.

03Crews

CrewAI · multi-role.

04SDKs

Labs + cloud natives.

How to choose in an interview

Decision tree from need to LlamaIndex, LangGraph, CrewAI, or provider SDKs.
Memorize this tree — it's enough for most system design prompts.
  • Knowledge only → RAG stack (RAG + LlamaIndex/Haystack); agent framework optional.
  • Unknown action sequence + side effects → tool loop; LangGraph or a provider Agents SDK.
  • Approvals, resume, long runs → graph/workflow with checkpoints (LangGraph + queue/Temporal).
  • Explicit specialist roles for a demo → CrewAI / supervisor pattern; harden before prod.
  • One model vendor forever → that vendor's agent SDK (less glue, more lock-in).
  • TypeScript product surface → Mastra or Vercel AI SDK + your own orchestration discipline.
Layers: framework, your app, platform, ops.
Frameworks don't replace auth, sandboxes, eval, or durable infrastructure.

LangGraph (and LangChain)

LangGraph is the production-oriented control plane in the LangChain ecosystem: you define a state graph (nodes + edges), run an agent/tool cycle as graph steps, and optionally checkpoint state for resume and human-in-the-loop interrupts. Deep Agents is the opinionated harness on top (planning, filesystem, subagents) — full Q&A in Deep Agents, LangGraph & LangChain.

START to agent to tools loop with HITL then END.
Explicit edges beat hidden magic for auditability and debugging.
  • Best for — stateful agents, branching, retries, HITL, multi-step tools with clear stop conditions.
  • Strengths — durable state, conditional routing, LangSmith/observability path, Python + JS/TS.
  • Trade-offs — steeper learning curve than "crew" APIs; easy to over-graph simple RAG.
  • LangChain — still useful for model/tool adapters and chains; for serious agents, prefer LangGraph's explicit graph over sprawling LCEL soup.

LlamaIndex and Haystack

LlamaIndex shines when the hard problem is data: connectors, chunking, indexes, query engines, and workflows that wrap retrieval + agents. Think document copilots, research over corpora, and RAG agents that need strong ingest.

  • Best for — RAG-first products, heterogeneous document sources, retrieval eval loops.
  • Agents — available, but don't pick LlamaIndex only to get a tool loop — pick it for the data plane.
  • Haystack — pipeline-first (nodes/pipelines), strong for production RAG in Python; good interview alternative when "modular pipelines" fit the story.

LlamaIndex mental model

Documents → Nodes (chunks) → Indexes → Retrievers → Query engines / Workflows. Agents sit on top of query engines. If your pain is ingest quality, start here; if your pain is HITL refunds, start with LangGraph.

CrewAI and AutoGen / AG2

CrewAI models work as a crew of role-playing agents (researcher, writer, reviewer) with sequential or hierarchical processes. Fast to demo multi-agent collaboration.

Crew with researcher, writer, reviewer, optional manager.
Role APIs match how PMs describe "AI teams" — validate with golden tasks before production.
  • Best for — prototypes, content/research pipelines with clear roles, stakeholder-friendly mental model.
  • Watch-outs — token cost, non-determinism, weaker "enterprise control" story than graphs unless you add your own guards.
  • AutoGen / AG2 — conversational multi-agent pioneer; still useful for research patterns. For new Microsoft-aligned work, prefer Microsoft Agent Framework (AutoGen + Semantic Kernel lineage consolidated).

Provider and cloud agent SDKs

Four cards: OpenAI Agents, Claude Agent, Google ADK, Microsoft Agent Framework.
First-party SDKs optimize for their ecosystem — great velocity, real lock-in.

OpenAI Agents SDK

Agents, tools, handoffs, guardrails, sessions, and first-party tracing. Lowest friction if you're already on OpenAI models and want a managed loop. Often provider-flexible via adapters, but the gravity stays OpenAI.

Claude Agent SDK (Anthropic)

Anthropic-first agent runtime emphasizing tool use, longer autonomous runs, and tightly integrated Claude capabilities (files/code/computer-use style features depending on product). Prefer when Claude is the model of record.

Google ADK

Code-first agents aimed at Gemini / Vertex paths, with multi-agent and emerging interoperability (e.g. A2A-style protocols). Natural default for GCP shops.

Microsoft Agent Framework

Enterprise/.NET + Python successor direction from Semantic Kernel and AutoGen: Azure alignment, MCP/A2A-style integrations, org-friendly packaging. Default when the interviewer says "we're a Microsoft stack."

Typed Python and TypeScript options

Pydantic AI

FastAPI-like DX: typed dependencies, structured outputs, agent definitions that feel like application code. Strong when your backend is already Pydantic/FastAPI and you want agents that don't fight the type system.

Mastra / Vercel AI SDK

Vercel AI SDK dominates streaming UI + model plumbing in TS/Next. Mastra pushes further into agent workflows for TypeScript products. Interview angle: frontend/streaming excellence ≠ durable multi-day agents — you may still need a workflow engine.

Others worth naming

  • smolagents (HF) — minimal agents; great for learning loops, not a full platform.
  • Agno / Phidata-line tools — agent-centric, data/tool oriented; mention if relevant to the prompt.
  • Haystack — covered above for RAG pipelines.
  • Raw provider SDKs — still valid for a single tool-calling loop with your own max-steps.

Comparison cheat sheet

Scatter of frameworks on control vs multi-agent axes.
Relative placement for interviews — not a scientific benchmark.
FrameworkControl modelSweet spotWatch-out
LangGraphGraph + stateProduction agents, HITLOverkill for FAQ RAG
LlamaIndexIndexes + workflowsDocument / RAG agentsNot primarily a crew runtime
HaystackPipelinesModular RAG prodLess "agent buzz" API
CrewAIRoles / crewMulti-role prototypesCost + eval discipline
OpenAI Agents SDKAgent + handoffsOpenAI-centric appsEcosystem gravity
Claude Agent SDKTool-centric agentClaude-first autonomyVendor path
Google ADKAgents + toolsGCP / GeminiVendor path
MS Agent FrameworkEnterprise agents.NET / AzureStack coupling
Pydantic AITyped agentsPython servicesYounger ecosystem than LC
Mastra / AI SDKTS workflows / UINext.js productsPair with durable backend

What no framework gives you for free

  • AuthZ / tenancy — tool credentials per user; never a shared admin key.
  • Sandboxing — code/shell/browser isolation and egress allowlists.
  • Idempotency — retries must not double-charge (see durable patterns in Agentic architectures).
  • Eval suite — golden goals, trajectory metrics, safety cases.
  • Cost controls — max steps, model tiers, kill switches.
  • Prompt/version ops — templates as code; see Prompt engineering.

Thin adapter layer

Wrap framework calls behind your interfaces (AgentRuntime.run(goal, tools, policy)). Business tools stay framework-agnostic plain functions. When LangGraph/CrewAI/API shifts, you swap the adapter — not 40 product call sites.

Version pinning and upgrades

Pin framework + model versions in CI; run golden trajectories before bumping. Framework 0.x releases break graphs and tool schemas — treat upgrades like dependency major versions, with rollback flags.

Migration and lock-in playbook

Assume you will change frameworks once. Design for that day:

  1. Week 0 — extract tools as plain functions; no framework imports inside domain code.
  2. Week 1 — define AgentRuntime interface: run(goal, tools, policy) → RunResult.
  3. Week 2 — implement adapter for current framework; golden trajectories pass.
  4. Cutover — feature flag 5% traffic to new orchestrator; compare task success and $/run.
  5. Exit — delete old adapter when metrics match for 2 weeks.
Lock-in vectorRiskMitigation
Prompt format proprietaryHard to switch modelsStore prompts as your templates; map at edge
Trace SaaS onlyBlind without vendorOTel export required
Tool schema DSLRewrite all toolsJSON Schema / OpenAPI as source of truth
Checkpoint formatCan't resume old runsOwn run store; framework checkpoint optional

In your interview

Prompts: "Which framework would you use?" "LangChain vs LlamaIndex?" "Design a multi-agent research system" "We're on Azure/GCP/OpenAI — what do you pick?"

What to say out loud

"I'd pick the control model first. For a durable support agent with refunds, LangGraph with checkpoints and HITL before write tools — or the OpenAI Agents SDK if we're standardized on OpenAI and want faster delivery. Knowledge answers go through RAG with LlamaIndex-style ingest. I wouldn't start with CrewAI for production money movement; I'd use roles later if evals show a single agent thrashing. Either way we own tool ACL, sandboxes, max steps, and a golden-task suite."

  • Name 2 frameworks max with a reason — not a laundry list.
  • Tie the choice to stack constraints (language, cloud, model vendor).
  • Admit trade-offs (speed vs control vs lock-in).
  • Separate RAG libraries from agent runtimes.
  • Mention what you'd build if frameworks disappeared tomorrow (plain tool loop).

Cost and delivery levers

Extended decision matrix

ConstraintPickWhy
Need HITL + resume in PythonLangGraph (+ Temporal if days-long)Checkpoints, interrupts
Docs/PDF hellLlamaIndex / HaystackData plane strength
Stakeholder demo next weekCrewAI or OpenAI Agents SDKFast roles / handoffs
All-in on AnthropicClaude Agent SDKEcosystem fit
GCP / GeminiGoogle ADKVertex path
.NET / AzureMS Agent FrameworkOrg standard
FastAPI typed servicesPydantic AIDX + schemas
Next.js streaming UIVercel AI SDK ± MastraFrontend path
Strict multi-cloud + complianceThin custom loop + TemporalMinimal lock-in

Framework exit drill

Failure modes to mention

Call out at least one dependency failure (DB down, cache stampede, queue lag, region outage) and your mitigation (timeouts, retries with jitter, degraded mode, circuit breaker).

Interview Q&A by level

Practice saying these out loud for agentic frameworks. Interviewers grade clarity and judgment more than buzzwords.

Interview takeaway

Match depth to the bar: define → trade off → operate. Don't dump principal answers in an entry-level screen.

Wrapping up

Agentic frameworks are accelerators for known patterns: graphs, crews, RAG workflows, and vendor agent loops. Learn the landscape well enough to choose deliberately — then spend your interview capital on tools, safety, durability, and measurement.

Continue with Deep Agents, LangGraph & LangChain — Q&A, Agentic patterns, Agentic architectures, RAG, Prompt engineering, Vector databases, and Key technologies.

← Lattice