Senior Full Stack Engineer - Enterprise Platform
About the job
- At least 5 years of full-stack experience with production exposure to authentication and authorization.
- Experience integrating an enterprise identity provider and understanding the complexities beyond its documentation.
- Fluency with TypeScript and React for front-end development, Python for back-end development, and cloud identity primitives such as Cognito or an equivalent.
- Experience with SAML 2.0, OIDC federation, SCIM provisioning, and enterprise identity providers such as Entra ID, Okta, and Google Workspace.
- Build enterprise identity end to end, including SAML 2.0 and OIDC federation and SCIM provisioning against Entra ID, Okta, and Google Workspace, integrated with the organization and seat model.
- Own the tenancy layer covering organizations, seats, roles, entitlements, and shared credit pools, and make it self-service for administrators.
- Build audit and administration surfaces including access logs, session policy, and de-provisioning that revokes access.
- Partner with forward-deployed and customer-success engineers on customer rollouts.
- Own the enterprise single sign-on initiative from architecture through rollout.
- Handle migrations, edge cases, and login failure scenarios.
About the company
NinjaTech AI builds an autonomous AI personal assistant called Ninja that handles research, writing, file analysis, coding, image generation, and meeting scheduling for busy professionals. Ninja takes conversations in chat, voice, or video with a customizable avatar and uses its own large language models plus a proprietary decision engine to plan and execute tasks end-to-end, turning requests into completed work within a SaaS platform. It stands apart by delivering full task execution rather than just suggestions, integrating planning, action, and communication in one system, supported by a team with backgrounds at Google, Meta, and AWS. Its goal is to democratize access to executive‑level assistance, helping more professionals save time and focus on their core responsibilities.
Company Size
11-50
Company Stage
Seed
Total Funding
$5M
Headquarters
Los Altos, California
Founded
2022
Get referred to NinjaTech AI
See people who can refer or advise you
Simplify's Take
What believers are saying
- Infosys invested August 4, 2026, opening large-enterprise distribution and delivery channels.
- September 2026 pricing promises about 10x lower costs than frontier-model deployments.
- Microsoft, AWS, and Fireworks AI partnerships reduce infrastructure friction for enterprise adoption.
What critics are saying
- September 2026 launch still lacks public customer logos, revenue, or usage disclosures.
- Open-weight dependence on DeepSeek V4.1 Flash invites model-quality and supplier-switching risk.
- One-year Infosys exclusivity and 1–50 employees constrain delivery if enterprise demand spikes.
What makes NinjaTech AI unique
- SuperNinja Enterprise, launched September 28, 2026, bundles software, GPUs, and inference.
- NinjaTech runs inside customer clouds, keeping data and IP inside enterprise boundaries.
- Babak Pahlavan’s Google-acquired CleverSense pedigree and 2026 Infosys partnership sharpen enterprise credibility.
Help us improve and share your feedback! Did you find this helpful?
Benefits
Health Insurance
Dental Insurance
Vision Insurance
Company Equity
Performance Bonus
Growth & Insights and Company News
6 month growth
-9%1 year growth
-3%2 year growth
0%NinjaTech launches fixed-price AI platform for enterprises. Tue, 13th Oct 2026 (Today) NinjaTech AI has launched SuperNinja Enterprise, an AI workforce platform that companies can run in their own cloud environments. It is sold at a fixed annual price that includes computing capacity. The Palo Alto-based company says the platform lets businesses deploy AI agents inside their own cloud or data centre rather than relying on an external service. It uses open-weight models and also supports models from Anthropic and OpenAI. SuperNinja Enterprise combines software, model-serving infrastructure and GPU-backed inference capacity under one contract. Customers can deploy the system on single-tenant capacity through Microsoft with Fireworks AI or AWS, while organisations with stricter security requirements can choose a licensed deployment in an air-gapped environment using their own hardware. The launch comes as companies look to move beyond small AI trials while keeping tighter control over spending and data handling. Usage-based pricing has made it difficult for many businesses to forecast costs as AI agents take on longer-running tasks and consume more computing resources than simple chatbot interactions. Cost model NinjaTech says organisations running open-weight models on reserved capacity can reduce end-to-end costs to about one-tenth of comparable deployments based on frontier lab models. The company is positioning the offer as an alternative to per-token billing, with annual pricing designed to stay stable as usage rises. That pricing model addresses a central concern for large companies testing AI tools at scale. Businesses often have to secure GPU capacity, deploy model-serving systems and source applications through separate suppliers, adding cost uncertainty and procurement complexity. Packages are sized for 100, 500 or 1,000 AI employees. The platform also integrates with Slack and Microsoft Teams, allowing workers to use the system through tools already common in the workplace. Data control NinjaTech is also emphasising data residency and separation. Enterprise data stays within the customer's own environment and is not pooled to train external models, a point likely to matter in sectors handling regulated or commercially sensitive information. "With SuperNinja Enterprise, your CISO can rest assured data never goes outside your own walls," said Babak Pahlavan, Chief Executive Officer of NinjaTech AI. "Your CFO gets an AI bill that doesn't move. And everyone sleeps well, knowing you can scale AI without waking up to skyrocketing costs." For the first time, NinjaTech is supplying the GPU and inference capacity for the platform rather than leaving customers to assemble those components themselves. In practice, that means buyers get the AI software and the underlying computing arrangement under one commercial agreement. Partners involved Healthcare customers will work with Optimum HealthcareIT for integration, while other enterprise customers will work with Infosys. Infosys is also described by NinjaTech as an investor and implementation partner. The product keeps model choice open, allowing customers to move workflows between open-weight systems and proprietary models. That approach reflects a broader enterprise effort to avoid dependence on a single model provider as pricing, performance and policy terms continue to shift across the AI market. NinjaTech says many enterprise AI projects remain stuck at the pilot stage because finance teams struggle to predict usage-based charges and security teams resist sending sensitive workloads to outside providers. It argues that placing the system inside a customer-controlled environment with fixed annual pricing removes both barriers. "Most vendors keep AI on a usage meter," said Pahlavan. "We give enterprises predictable capacity and costs, so adoption can spread across the organisation instead of stopping at a pilot."
NinjaTech AI, an early-stage startup backed by Amazon and led by a former Google product management executive, has unveiled a new product designed to help large companies forecast the costs of running agents continuously for long-running tasks. The tool aims to eliminate uncertainty in budgeting for agent operations that run around the clock. The startup's solution addresses growing concerns about the unpredictable expenses associated with deploying AI agents at scale. By providing more accurate cost projections, NinjaTech AI seeks to help enterprises better plan their AI infrastructure investments and avoid unexpected bills from agent deployments.
NinjaTech AI launches SuperNinja Enterprise, a Turnkey AI workforce platform enterprises own and run Unmetered on open-weight model. Built to end runaway per-token AI costs, the platform deploys into the customer's own cloud and data centers, so data never leaves their control. Sep 28, 2026 Prev Next 1 of 43,225 NinjaTech AI launched SuperNinja Enterprise, an AI workforce platform that puts AI employees to work around the clock inside the customer's own cloud, on a fixed annual bill, with the computing capacity delivered under the same contract. It runs open-weight models and keeps enterprise data inside the customer's boundary. For the first time, NinjaTech AI is also supplying the GPU and inference capacity the platform runs on. Customers get one product that combines the AI software, model-serving infrastructure and compute, rather than assembling those pieces from separate vendors. "With SuperNinja Enterprise, your CISO can rest assured data never goes outside your own walls," said Babak Pahlavan, CEO of NinjaTech AI. "Your CFO gets an AI bill that doesn't move. And everyone sleeps well, knowing you can scale AI without waking up to skyrocketing costs." What Sets SuperNinja Enterprise Apart * Lower cost than frontier models: Total cost is about 10x lower, running open-weight models. * Turnkey capacity: GPUs, inference and software in one contract, with no infrastructure vendor to assemble first. * Unmetered open-weight compute: AI reaches every employee, not stopping at a rationed pilot. * AI workforce that enterprises own: AI employee production and outputs remain in the enterprise's own tenant. * Enterprise data is not pooled: It never trains any outside models. * Fixed annual pricing: Consistent annual pricing that holds as usage increases. * 24/7: AI employees that take a goal and finish the work, running unattended around the clock. * Swappable models: The freedom of never being locked into one AI lab. * Works in the tools enterprises use: Integration into Slack and Microsoft Teams. * Expandable on-demand: Packages sized for 100, 500 or 1,000 AI employees. Cost and Data Concerns Limit Enterprise AI Adoption Many enterprise AI initiatives stall for two reasons. Teams cannot forecast usage-based pricing, and they can't hand their most sensitive work to an outside AI service. A fixed annual bill inside the customer's own environment answers both. AI agents have advanced enough to handle long, unattended jobs, but that means they now burn far more tokens than a chat interface ever did. But while token prices are generally falling, enterprise AI bills keep climbing. That has kept most enterprise AI deployments stuck in the pilot stage instead of scaling across organizations to drive stronger results. Model choice stays open. The platform supports Anthropic and OpenAI alongside open-weight alternatives, so customers can move a workflow to the best model for the job without committing to a single AI lab. Single-Contract Deployment Model Deploying enterprise AI today usually means securing GPU capacity, standing up model-serving infrastructure and deploying applications across separate vendors and contracts. The offering combines all three into one. NinjaTech AI delivers GPU and inference capacity through Microsoft with Fireworks AI or AWS, on single-tenant capacity reserved for the customer. Neither side buys hardware. Organizations that run open-weight models on that reserved capacity see end-to-end costs about 10 times lower than comparable deployments on frontier lab models. Customers whose most sensitive work cannot touch an external network can instead choose a licensed deployment inside their own air-gapped environment, where they supply the hardware. NinjaTech delivers the dedicated capacity above end-to-end. Platform and implementation arrive in one contract at one price. Healthcare customers work with NinjaTech AI integration specialist Optimum HealthcareIT, and other enterprises work with Infosys. SuperNinja Enterprise is available now. Enterprises can start with a fixed-scope pilot that puts AI employees on real work in days, then expand into an annual capacity contract. "Most vendors keep AI on a usage meter," Pahlavan said. "We give enterprises predictable capacity and costs, so adoption can spread across the organization instead of stopping at a pilot."
Infosys and NinjaTech AI-Building the Enterprise AI Workforce, Together. By The Ninja AI Team August 4, 2026 The Infosys Innovation Fund has made a strategic investment in NinjaTech AI, and the two companies are partnering to bring an enterprise-grade AI workforce to large, complex organizations. The belief behind it is simple: AI should amplify human potential, and enterprises should get that power on their own terms. Why Infosys invested. NinjaTech has built an autonomous, self-evolving AI workforce that works alongside every employee, inside the enterprise's own environment, and is measured by business outcomes rather than demos. It answers the three questions that decide enterprise AI: how to deploy without handing over your data, how to pay without a bill that balloons with every user, and how to reach every employee rather than one team. The four capabilities below are how. 24/7, self-evolving AI employees. NinjaTech's AI employees run complex, multi-step work from start to finish, around the clock, and get better on their own from the work they do. Because most enterprise value is locked in long processes that cross teams, an AI employee that owns a whole process keeps compounding in value instead of leveling off, and capacity grows with demand rather than headcount. Deployed in your own cloud. NinjaTech runs inside the customer's own cloud accounts on AWS, Microsoft Azure, or Google Cloud, on infrastructure the customer owns and controls. For regulated industries, and any company that treats its data as a strategic asset, that keeps data, access, and governance inside the compliance boundary the company already trusts, and sensitive information never leaves the perimeter. Your models, or unlimited usage. NinjaTech works with the best frontier models, including those from OpenAI and Anthropic, and can also run unmetered on open-weight models such as Moonshot AI's Kimi K3 on the customer's own GPUs. The highest-value work can use the strongest proprietary models, while high-volume work runs on owned hardware with no per-seat or per-token metering. That turns an unpredictable, usage-based bill into a fixed cost, which is what makes it realistic to give AI to an entire workforce. Human + AI, in the flow of work. From Slack and Microsoft Teams, any employee can bring in as many AI teammates as they need, hand off work, and stay in the loop. By meeting people inside the tools they already use, NinjaTech turns the AI workforce into something they reach for by default, so it reaches the whole organization instead of a single pilot team. What Infosys and NinjaTech are building together. Infosys and NinjaTech are partnering to accelerate enterprise adoption of the AI workforce across both sales and delivery, and Infosys will be the exclusive professional services partner for NinjaTech AI for at least one year. Infosys brings trusted relationships with the world's largest enterprises, deep industry expertise, and the scale to deliver responsibly; NinjaTech brings the product. Together NinjaTech AI, Inc. can meet a large enterprise where it is, design the workforce around its real processes, and stand it up quickly and securely, with a single accountable partnership from strategy through to production. The opportunity ahead. Every major enterprise wants to put AI to work safely, affordably, and at scale, and few offerings deliver all three at once. NinjaTech's technology with Infosys's reach and delivery is built for exactly that gap. NinjaTech AI, Inc. is proud to have the Infosys Innovation Fund as an investor and Infosys as its services partner, and NinjaTech AI, Inc. is just getting started. A workforce, not a tool. A platform for AI employees that work around the clock, self-evolve to your business, finishing real work end to end from Slack, Teams, and WhatsApp...
NinjaTech AI Inc. in July launched SuperNinja, which operates on a computer in the same way human workers do, according to the company's vice president of product, Divya Chirasani.