Skip to content

cryptography-engineering.md

Latest commit

 

History

History
296 lines (239 loc) · 15.1 KB

cryptography-engineering.md

File metadata and controls

296 lines (239 loc) · 15.1 KB

Cryptography Engineering Career Roadmap

📘 Recommended study plans: Common Skills · Product Security · Secure Code Review.

Cryptography Engineering is a specialized, math-flavored, deeply technical corner of cybersecurity. It is one of the smallest career tracks by headcount, but consistently among the best paid because qualified people are rare and the cost of mistakes is catastrophic (think Heartbleed, Dual_EC_DRBG, ROBOT, Logjam, Padding Oracle). Most companies don't invent cryptography — but every product company needs at least a few people who can integrate, review, and operate it correctly.

Important distinction:

  • Cryptographer / Cryptanalyst (research) — invents and breaks algorithms. PhD-level math, mostly academia / NIST / national labs / vendors like Microsoft Research, Cloudflare Research, IBM Research.
  • Cryptography Engineer (this roadmap) — applies, integrates, reviews, and runs cryptographic systems. Strong CS + math + security background; does not require a PhD.

This roadmap focuses on Cryptography Engineering. The research path is briefly noted at the end.

Who is this for?

  • AppSec / software engineers fascinated by crypto bugs and protocol design
  • Security engineers in BFSI, payments, identity, blockchain, or messaging
  • Cloud security engineers handling KMS / HSM / key lifecycle at scale
  • People who enjoyed the cryptography sections of CTFs and want to do that for a living
  • Folks aiming for niche, hard-to-replace, well-compensated roles

Pre-requisites (foundation)

  1. Strong programming — at least one of C, Rust, Go, Java, or Python (with crypto libs)
  2. Computer architecture basics — endianness, memory, timing, side channels
  3. Discrete math basics — modular arithmetic, group theory at high level, basic number theory
  4. Linear algebra and probability (high-school+ level enough to start)
  5. Networking + TLS at the level of Bulletproof TLS and PKI (Ivan Ristić)
  6. Solid understanding of common AppSec issues — you'll review code that uses crypto wrong
  7. Comfort reading RFCs, NIST SPs, and academic papers

What to learn (core crypto knowledge)

Symmetric crypto

  • Block ciphers — AES (modes: ECB never, CBC, CTR, GCM, GCM-SIV, XTS for disk)
  • Stream ciphers — ChaCha20, ChaCha20-Poly1305 (AEAD)
  • Hash functions — SHA-2, SHA-3, BLAKE2/3
  • MACs — HMAC, KMAC, Poly1305
  • KDFs — HKDF, PBKDF2, scrypt, Argon2 (password hashing)
  • AEAD — why nonce reuse breaks GCM; misuse-resistant constructions

Asymmetric crypto

  • RSA (PKCS#1 v1.5 vs OAEP, PSS); why textbook RSA is broken
  • Elliptic curves — P-256, P-384, Curve25519, Ed25519, secp256k1
  • Diffie-Hellman, ECDH, X25519
  • Signatures — ECDSA, EdDSA, RSA-PSS
  • Key exchange and forward secrecy

Protocols & systems

  • TLS 1.2 vs 1.3 in depth — handshake, ciphersuites, key schedule
  • Noise Protocol Framework (used by WireGuard, WhatsApp)
  • Signal Protocol (X3DH + Double Ratchet)
  • Kerberos and PKI (X.509, OCSP, CT logs)
  • JWT / JWS / JWE / JWK — and why JWT is mis-designed in many ways
  • PASETO, Macaroons, Biscuit — modern token alternatives
  • OAuth 2.1 + OIDC crypto pieces (DPoP, mTLS-bound tokens)

Key management

  • KMS — AWS KMS, GCP KMS, Azure Key Vault, HashiCorp Vault, GCP Confidential Space
  • HSM — Thales Luna, AWS CloudHSM, Entrust nShield, YubiHSM
  • Standards — FIPS 140-2 / 140-3, Common Criteria (CC EAL)
  • Envelope encryption, key wrapping, key rotation, BYOK / HYOK
  • Secrets management vs key management (different problems)

Post-Quantum Cryptography (PQC) — now mainstream

This is the big shift in 2024–2030. NIST finalized:

  • ML-KEM (Kyber) — key encapsulation
  • ML-DSA (Dilithium) — digital signatures
  • SLH-DSA (SPHINCS+) — stateless hash-based signatures
  • FN-DSA (Falcon) — short-signature alternative

Topics to know:

  • Hybrid key exchange (ECDH + ML-KEM)
  • Crypto agility — designing systems so algorithms can be swapped
  • "Harvest now, decrypt later" threat model
  • CNSA 2.0 timelines, BSI / ANSSI guidance
  • TLS 1.3 PQ — Cloudflare, Google, AWS rollouts

Side-channel & implementation security

  • Timing attacks — constant-time programming, crypto/subtle style APIs
  • Cache attacks (FLUSH+RELOAD, PRIME+PROBE)
  • Power / EM analysis (relevant for embedded / IoT / HSM design)
  • Fault injection
  • Speculative execution (Spectre / Meltdown family)
  • Why never roll your own crypto — and what integrating it correctly looks like

Specialized areas

  • Privacy-Enhancing Technologies (PETs) — differential privacy, secure multiparty computation, homomorphic encryption (FHE), zero-knowledge proofs (zk-SNARK / zk-STARK)
  • Confidential Computing — Intel SGX, AMD SEV-SNP, ARM CCA, Nitro Enclaves
  • Blockchain crypto — secp256k1, BLS, Schnorr, threshold signatures, MPC wallets

Career ladder

Entry — Junior Cryptography / Applied Security Engineer (0–2 yrs in crypto, 2–4 yrs total)

Most people enter from AppSec, software engineering, or backend.

Typical work

  • Implement cryptographic features under guidance (mTLS, JWT signing, HMAC, payload encryption)
  • Code-review libraries' usage — catch ECB, hardcoded IVs, weak KDFs, custom crypto
  • Integrate KMS / Vault into services
  • Write threat models for crypto-using features

Skills

  • Strong language (Go / Rust / Java / C / Python) + a major crypto lib (libsodium, BoringSSL, Tink, ring, RustCrypto)
  • Solid TLS 1.2/1.3, certificate handling
  • Comfortable with at least AWS KMS or GCP KMS or Vault
  • Cryptography I (Boneh, Coursera) completed

Mid — Cryptography Engineer (3–6 yrs total)

Typical work

  • Design crypto components — token formats, encrypted storage, signed audit logs
  • Build internal libraries / SDKs that wrap primitives so other engineers can't misuse them
  • Lead crypto reviews across the org
  • Drive HSM / KMS architecture decisions
  • Run incident response on crypto-related findings (downgrade attacks, weak ciphersuites, expired certs)

Skills

  • Deep TLS, PKI, key management
  • Read RFCs and NIST SPs fluently
  • Build constant-time, side-channel-aware code
  • Strong AppSec foundation
  • Familiarity with PQC roadmap and crypto agility patterns

Senior — Senior Cryptography Engineer (6–10 yrs)

Typical work

  • Own crypto across an entire product line / platform
  • Design end-to-end-encrypted features (E2EE messaging, encrypted backups, sealed sender)
  • Drive PQC migration program org-wide
  • Speak at conferences (Real World Crypto, USENIX Security, Black Hat)
  • Mentor mid-level engineers; set crypto coding standards

Skills

  • Protocol design experience (Noise, Signal-style ratchets, custom AEAD constructions when justified)
  • Confidential Computing or HSM-deep work
  • Public-facing voice — blog posts, RFC contributions, open-source maintenance

Staff / Principal — Principal Cryptography Engineer (10+ yrs)

Typical work

  • Define crypto strategy across multiple BUs
  • Drive standardization with external bodies (IETF, NIST, CA/B Forum)
  • Architect E2EE / PQC / Confidential Compute platforms
  • Be the org's "Court of Last Appeal" on crypto questions

Where these jobs exist

  • Hyperscalers (AWS / Google / Microsoft / Apple)
  • Payment / fintech (Stripe, Square, Visa, Mastercard, banks)
  • Messaging / privacy (Signal, WhatsApp, Apple)
  • Cloud-native security vendors (Cloudflare, HashiCorp, Hashi, Sigstore)
  • Web3 / blockchain infrastructure (Coinbase, Chainlink, zk-rollup teams)
  • Government / defense / national labs

Specialization branches

PQC migration specialist

  • Help orgs inventory cryptographic assets, pilot hybrid TLS, migrate signing systems
  • Hot in 2025–2030; many BFSI / government mandates
  • Tools: open-quantum-safe (OQS), liboqs, Cloudflare's CIRCL

Confidential Computing engineer

  • SGX / SEV-SNP / TDX / Nitro Enclaves
  • Attestation, sealed storage, multi-party AI workloads on encrypted data
  • Strong overlap with AI security (private inference, federated learning)

Blockchain / Web3 cryptography

  • Threshold signatures (TSS), MPC wallets, BLS aggregation, zero-knowledge circuits
  • Custodian / exchange / L2 infra

Privacy engineering

  • Differential privacy, federated learning, FHE, MPC, zk-proofs for compliance and analytics
  • Hybrid GRC + Crypto + AppSec role

Cryptographer / Cryptanalyst (research)

  • PhD path; algorithm design, formal analysis, attack publication
  • Research labs (Microsoft Research, IBM, Cloudflare, Inria, ENS, MPI)

Career paths from Cryptography Engineering

                AppSec / SWE / Backend Engineer
                              │
                              ▼
                Junior Cryptography Engineer
                              │
                              ▼
                  Cryptography Engineer
              ┌───────────────┼────────────────┐
              ▼               ▼                ▼
        Senior Crypto    PQC Migration    Confidential
         Engineer          Specialist     Computing Eng
              │                │                │
              └────────┬───────┴────────┬───────┘
                       ▼                ▼
              Principal Cryptography  Privacy Engineer
                  Engineer            (PETs / FHE / DP)
                       │                       │
                       ▼                       ▼
              Distinguished Engineer    Web3 / Blockchain
              / Security Architect      Crypto Lead
                       │
                       ▼
                Researcher / Cryptographer
                  (PhD-track / labs)

Lateral pivots from / into Cryptography Engineering

  • AppSec → Crypto Engineering — the most common path
  • Backend SWE → Crypto Engineering — if you've worked on auth / tokens / payment systems
  • Crypto → Security Architect — protocol design experience translates well
  • Crypto → Confidential Computing / Privacy Engineering
  • Crypto → AI Security — secure aggregation, federated learning, private inference

AI-augmented Cryptography Engineering (2025+)

AI is especially dangerous here — crypto is one of the fields where LLMs hallucinate plausibly wrong code most often.

Using AI carefully

  1. Code review prompts — paste a function and ask for misuse patterns; cross-check against the spec
  2. RFC / NIST SP summarization — great for first pass; never quote verbatim without re-reading
  3. Generating test vectors — ask AI for KAT-style inputs from RFC appendices to seed property-based tests
  4. Formal threat model first drafts — STRIDE-style around a crypto component
  5. PQC migration discovery — feed source / dependency manifests to AI to flag legacy crypto

Red flags

  1. AI-generated crypto code — almost always wrong on padding, IV handling, constant-time, or AEAD usage. Treat as a hint, never as final code.
  2. Side-channel reasoning — LLMs are weak here; involve a human expert.
  3. Custom protocols — never let an LLM design one for you.

Securing AI workloads

  • Confidential inference (Nitro Enclaves, Intel TDX) — direct overlap with this track
  • Secure aggregation in federated learning (Bonawitz et al.)
  • Watermarking + signed model artifacts (Sigstore for models)

See: AI Security Career Roadmap · GenAI Security Study Plan

Recommended tools / libraries to know

  • High-level libs: libsodium, Google Tink, AWS Encryption SDK, age
  • Low-level libs: BoringSSL, OpenSSL, ring (Rust), RustCrypto, Bouncy Castle
  • PQC: liboqs / open-quantum-safe, CIRCL, pq-crystals
  • KMS / HSM: AWS KMS + CloudHSM, GCP KMS, Azure Key Vault, HashiCorp Vault, Thales Luna, YubiHSM, SoftHSM (lab)
  • Test / fuzz: Wycheproof test vectors, AFL/libFuzzer for crypto code
  • Confidential Computing: Open Enclave SDK, Gramine, Constellation, Microsoft CCF

Recommended labs / resources

  • Cryptopals Crypto Challenges — the gold standard intro
  • CryptoHack — modern, gamified
  • Real World Crypto conference YouTube channel
  • IETF mailing lists — TLS WG, CFRG
  • NIST PQC project pages and standards (FIPS 203 / 204 / 205)
  • A Graduate Course in Applied Cryptography — Boneh & Shoup (free PDF)

Recommended books

  • Serious Cryptography — Jean-Philippe Aumasson (best modern intro)
  • Cryptography Engineering — Ferguson, Schneier, Kohno
  • Bulletproof TLS and PKI — Ivan Ristić
  • Real-World Cryptography — David Wong
  • The Code Book — Simon Singh (history / motivation)
  • Handbook of Applied Cryptography — Menezes, van Oorschot, Vanstone (free, dense, reference)
  • Post-Quantum Cryptography — Bernstein, Buchmann, Dahmen (eds.)

Certifications (limited, but useful)

  • (ISC)² CISSP — broad, useful for senior roles
  • EC-Council ECES — Encryption Specialist (entry-level signal)
  • Cloud KMS specialty — vendor-specific (AWS Security Specialty, Google PCSE)
  • Practical DevSecOps CDP / CDE — for the AppSec + Crypto integration angle
  • No widely-respected dedicated crypto-engineer cert exists. Public open-source contribution + a strong blog matter more.

Recommended creators / blogs to follow

  • A Few Thoughts on Cryptographic Engineering — Matthew Green
  • Cloudflare Research blog
  • Trail of Bits blog (crypto reviews are gold)
  • Filippo Valsorda (filippo.io) — Go crypto + PQC commentary
  • NCC Group Cryptography Services — published reviews
  • Real World Crypto YouTube channel

Common pitfalls (what employers want you NOT to do)

  1. Roll your own crypto / "improve" a standard
  2. Use ECB anywhere
  3. Use the same nonce twice with GCM
  4. Use MD5 / SHA-1 / RC4 / DES for anything new
  5. Confuse encryption with authentication (use AEAD)
  6. Use == to compare MACs (constant-time required)
  7. Bake algorithm choice into a binary format with no version field (no crypto agility)
  8. Trust JWT defaults blindly (none-alg, alg confusion, key confusion)
  9. Hardcode keys in source / mobile apps / containers
  10. Ship without an inventory of where crypto is used (you can't migrate to PQC if you can't find it)

Next step

Pair this roadmap with the AppSec / Software Security roadmap — most crypto engineering jobs are filled by strong AppSec engineers who specialized. If you want the math-research path, plan for a PhD; if you want the engineering path, build a public portfolio of crypto reviews + open-source contributions.

Companion JDs: cryptography roles often live inside Software Security JDs and Security Architect JDs.