Secrets Management, Machine vs Human Identities & Video AI — With Suresh Sathyamurthy (Akeyless)

Secrets Management, Machine vs Human Identities & Video AI — With Suresh Sathyamurthy (Akeyless)

Jump to
Heading H2

Podcast Transcription

Got it. I turned that dialogue into a clean, skimmable article with only the useful timestamps as H2s. Neutral tone, no jargon soup.

Managing Machine Identities at Scale: Why Secrets Management Has to Evolve

Summary

The conversation with Akilus digs into why identity security is shifting from humans to machines, what “secrets management without the management” means, and how to build brand trust when attention spans are microscopic. Core ideas: machine identities now dwarf human ones, static secrets are a breach magnet, cloud-native SaaS plus customer-controlled keys changes the risk model, and short video with sharp hooks wins attention while evidence wins deals.

Key takeaways

  • For every 1 human identity there are ~45 machine identities in 2024, trending toward 100+ per human.
  • Static secrets in code and config files are the biggest current risk.
  • A unified, cloud-native platform with customer-controlled key fragments reduces overhead and blast radius.
  • Short video with clear hooks builds awareness; customer evidence maintains attention through long cycles.
  • Podcasts create pipeline by giving prospects a credible, long-form forum. Remote works fine.

00:00 Why Akilus exists

Identity security used to center on people. Now the volume and velocity of machine identities is the problem. Services, microservices, automated jobs and service accounts rely on certificates, credentials and keys. Those secrets are increasingly the cause of breaches. Akilus’ purpose is to prevent those breaches by managing the full lifecycle of non-human identities and the secrets they use.

00:03 Secrets management without the management

Traditional secrets tooling is often self-deployed in clusters. As identities grow, so does the burden: engineering time, SRE overhead, regional scaling, higher cost. Akilus offers a cloud-native SaaS model that offloads the infrastructure while keeping customers in control of their secrets.

How control is preserved

  • Distributed fragment cryptography: encryption key operations are split across clouds, and a fragment remains under the customer’s control.
  • Result: the provider cannot access the customer’s secrets, while the customer keeps the benefits of SaaS scale and availability.

00:05 Time to value

“Start at breakfast, done by lunch.” Because it is cloud-native and massively scalable, deployment and initial value are fast compared to self-managed vaults.

00:07 Who it is for

Best fit is multicloud and hybrid environments where teams juggle multiple cloud KMS offerings and on-prem systems. Akilus consolidates lifecycle management across those silos. Typical early adopters include financial services, retail, technology, and healthcare, but usage is horizontal.

00:10 Why customers pick Akilus

  1. Ease of use and deployment compared to self-managed or cloud-specific tools.
  2. Consolidation: one platform for discovery, management, protection and secure access for non-human identities.
  3. A practical path toward secretless machine authentication, evolving from static secrets to rotation, to dynamic just-in-time, to frameworks like SPIFFE.

00:12 The path off static secrets

Stage 1: Extract secrets from code and config into secure storage.
Stage 2: Automate rotation on a defined cadence.
Stage 3: Move to zero standing privileges with dynamic, just-in-time credentials.
Stage 4: Eliminate secrets where possible with secretless authentication (for example SPIFFE, JWT-based approaches).

Akilus supports all four stages so teams can progress from wherever they are today.

00:14 Machines vs humans: what changes

  • Machines do not log in like humans. They are ephemeral and dynamic.
  • Ownership is clear for people, murky for services.
  • Scale is the headline: ~45 machine identities per person in 2024, moving toward 100+ as agentic AI and machine-to-machine workflows expand.

00:16 Discovery to access: one lifecycle

Lifecycle starts with finding secrets and identities, then managing, protecting and enforcing secure machine-to-machine access. Today, many teams stitch together multiple tools. Akilus’ claim is end-to-end coverage in one platform.

00:17 Competitive set

  • Human IAM vendors extending into machines, but the fit breaks at scale and behavior.
  • Cloud provider secrets tools that are great in one cloud but awkward across multicloud and on-prem.
  • Self-deployed vaults that are costly to run and hard to scale.
    Akilus positions against all three with SaaS, multicloud reach, and customer-controlled key fragments.

00:19 Cost in a tight market

Self-managed clusters consume engineering, DevOps and SRE hours. SaaS reduces that overhead while maintaining security boundaries. The pitch: lower total cost without losing control.

00:21 Building trust upstream

Short attention spans favor two things: video and sharp hooks that get to the customer problem fast. Use those to pull prospects to a site that offers self-education paths: self-serve demos, request a demo, chat, or docs. Most buyers complete the majority of research before contacting sales, so invest above the funnel.

00:24 Staying top of mind during long cycles

Credibility compounds when peers validate you. Encourage customer stories in public forums, conferences and communities. Publish evidence, not opinions: savings delivered, breaches prevented, measurable improvements.

Search behavior is shifting toward LLMs, so structured, evidence-rich content is more likely to surface where people now ask questions.

00:27 Put experts on camera

If the topic is highly technical, let practitioners speak. Audiences want to hear from people who build and operate the systems. Podcasts are particularly strong for pipeline: authentic conversations with prospects create content and credibility at the same time. Remote recording is perfectly viable.

Subscribe to our Newsletter