1. Enterprise GenAI architecture and risk
Strong practical understanding of AI/ML and Generative AI, with deep architectural knowledge of enterprise GenAI / LLM systems (models, APIs, data flows, orchestration and infrastructure)
Hands-on experience with enterprise LLM APIs and orchestration frameworks
Ability to identify and manage GenAI-specific risks across the stack, including familiarity with OWASP Top 10 for Agentic Applications
2. Security assessment and red teaming for AI systems
Experience planning and executing penetration tests, audits and red team exercises for GenAI solutions
Proven ability to identify, evaluate and communicate AI/LLM-specific security and abuse risks (prompt injection, data exfiltration, model denial-of-service, unsafe tool use, jailbreaks, data leakage etc.)
Experience supporting security due diligence and architecture reviews with cross functional stakeholders (security, engineering, product, risk)
Experience working closely with GenAI, Product and Engineering teams to interpret findings and set risk based priorities
3. Systematic evaluation of AI behaviour and outputs
Ability to design and run structured evaluations of AI/LLM behaviour, including adversarial and misuse scenarios
Strong skills in reviewing and critically assessing AI-generated outputs for quality, clarity, safety, usability and user experience
Ability to produce accurate, consistent and actionable feedback that drives continuous improvement of models, prompts, guardrails and system design
Secondary skills:
1. Communication, stakeholder management and decision-making
Excellent written and verbal communication skills, including the ability to document findings clearly and influence decision makers
Strong stakeholder management across security, engineering, product and compliance
Ability to prioritise and make risk based decisions under time pressure
2. Analytical evaluation and critical thinking
Exceptional reading comprehension, attention to detail and observational skills




Ability to design, iterate and run repeatable evaluation processes and experiments
3. Security design, threat modelling and remediation leadership
Practical experience with threat modelling of AI/LLM systems and translating findings into concrete safety requirements and controls
Experience defining security and safety requirements for GenAI applications and agentic systems
Ability to manage external pentest / red team engagements and drive remediation to closure with engineering teams, including tracking and validating fixes
Good to have skills:
Strong software engineering background with hands-on coding experience in Python
Hands-on experience working with LLMs (fine tuning, configuration, guardrails, RAG, agents)
Demonstrated expertise in prompt engineering and context engineering with top tier LLMs (e.g. OpenAI GPT, Anthropic Claude, Google Gemini)
Ability to develop and automate evaluation frameworks to measure LLM and agent performance, including adversarial and red team style tests
Job Description:
We are looking for a Senior Security Engineer for a new GenAI Red Teaming initiative to conduct structured adversarial testing that identifies and mitigates security, safety and trust vulnerabilities in AI systems before they can be exploited. It involves simulating real-world threats to identify flaws like prompt injection, toxic outputs, model extraction, bias, knowledge risks, and hallucinations in large language models (LLMs), training data, or system outputs. To proactively identify and mitigate vulnerabilities, risks, and harmful behaviors in AI systems by simulating real-world threats and conducting holistic evaluations of models, deployment pipelines, and interactions. This enables us to strengthen AI security, reduce risks, and improve system resilience.
You will define and document requirements and the governance model for GenAI Red Teaming, support technical investigations to ensure transparent go / no-go recommendations, and capture key architectural decisions for model deployment and integration.

📌 - IT Security Engineer - - (Bengaluru)
🏢 TRIGENT SOFTWARE PRIVATE
📍 Bengaluru