CGI
Website:
cgi.com
Job details:
Position Description
Company Profile:
Founded in 1976, CGI is among the largest independent IT and business consulting services firms in the world. With 94,000 consultants and professionals across the globe, CGI delivers an end-to-end portfolio of capabilities, from strategic IT and business consulting to systems integration, managed IT and business process services and intellectual property solutions. CGI works with clients through a local relationship model complemented by a global delivery network that helps clients digitally transform their organizations and accelerate results. CGI Fiscal 2024 reported revenue is CA$14.68 billion and CGI shares are listed on the TSX (GIB.A) and the NYSE (GIB). Learn more at cgi.com.
Job Title: Senior Generative AI & Agentic Systems Engineer
Experience: 8+
Category: Software Development/ Engineering
Location: Bangalore
Position ID: J0226-2251
Employment Type: Full Time
Education Qualification Any Engineering background or and MCA / Any Graduation – Full Time only
Responsibilities:
- Design & build LLM-powered applications and agentic systems: Architect, develop, and productionize robust single- and multi-agent workflows (planner–executor, supervisor–worker, debate/reflection patterns) with tool use, memory, and long-running workflows.
- Implement Retrieval-Augmented Generation (RAG) at scale: Stand up high-accuracy retrieval pipelines (chunking, hybrid search, reranking, citation grounding) using vector databases and knowledge sources; ensure groundedness and traceability.
- Agent orchestration & tool integration: Leverage frameworks such as LangGraph, Semantic Kernel, LangChain, AutoGen, CrewAI, OpenAI Assistants, to connect agents to internal systems (REST/GraphQL, SQL, search, ticketing, email, CRM/ERP), and manage function calling & tool routing reliably.
- LLMOps & CI/CD for prompts and agents: Establish prompt versioning, experiment tracking, datasets for offline/online evaluation, canary rollouts, and automated regression testing for prompts, tools, and policies.
- Safety, security & governance by design: Implement guardrails (prompt-injection and jailbreak defenses, PII/PHI redaction, output filtering), data access controls, and policy compliance (e.g., SOC 2, GDPR); conduct adversarial testing and red-teaming.
- Inference performance & cost optimization: Optimize latency, throughput, and cost via semantic caching, batching, speculative decoding, quantization, and efficient serving (e.g., vLLM/TGI/Triton); manage token budgets and context strategies.
- Automated testing & evaluation: Build evaluation harnesses to measure helpfulness, harmlessness, groundedness, factuality, task success, and tool-use reliability; run A/B tests and continuous quality checks in production.
- Data engineering for GenAI: Curate and process enterprise data, build embeddings pipelines, maintain vector stores (e.g., FAISS/Milvus/Pinecone/Weaviate/Azure AI Search), and develop feature/knowledge stores for agents.
- Observability & reliability: Instrument traces, structured logs, and metrics (token usage, latency, hallucination rate, grounding score) using OpenTelemetry / LangSmith / W&B / Arize, and implement SLOs, alerts, and auto-remediation playbooks.
- Research → production: Evaluate and integrate new LLM families (open and hosted), multimodal models, routing strategies, and planning algorithms—harden them for production at enterprise scale.
- Documentation & collaboration: Produce clear design docs, threat models, and runbooks. Partner with product, security, data, and platform teams to shape roadmaps that deliver measurable business impact.
Foundational Skills
- Programming excellence (Python required; TypeScript/Node a plus) with strong software engineering fundamentals: testing, API design, modularity, and reliability.
- ML & statistics: Solid grasp of regression/classification, time series/forecasting, evaluation design, and bias/variance tradeoffs.
- ML pipelines & deployment: Experience building reproducible ML/LLM pipelines with CI/CD, artifact management, and automated tests.
- Cloud fluency: Building scalable AI services in at least one major cloud (Azure/AWS/GCP) with secure, automated deployments.
Specialized Skills (GenAI & Agentic Systems)
- LLMs & transformers: Strong knowledge of prompt engineering, function calling, tool use, context strategies, and model selection (open vs hosted).
- Agentic frameworks: Hands-on with LangGraph / Semantic Kernel / LangChain / AutoGen / CrewAI / OpenAI Assistants and multi-agent coordination patterns.
- RAG & retrieval: Expertise in chunking, hybrid/vector search, reranking, query transformation, grounding with citations, and domain adaptation.
- Evaluation & guardrails: Designing offline/online evals, dataset curation, hallucination detection, and safety policies (toxicity, PII, jailbreak resistance).
- Data engineering & big data: Proficiency with Spark/distributed compute, streaming (Kafka/PubSub), and data processing frameworks.
- Vector databases & search: Practical experience with FAISS/Milvus/Weaviate/Pinecone/Azure AI Search, schema design, indexing, and consistency models.
- MLOps & LLMOps: Prompt/version control, experiment tracking (W&B/MLflow), model registries, deployment automation, and observability.
- Optimization: Caching, batching, quantization, GPU inference, cost control, and throughput tuning.
- Multimodal AI: Integrating text with images/tables/code/structured data; building evaluators across modalities.
Qualifications & Requirements
- BSc/MSc/PhD in Computer Science, Data Science, or related field; 5+ years building cloud-based ML/AI systems in production (incl. solution architecture/design).
- 8+ years hands-on Python for AI/ML; strong with data/ML libraries (e.g., PyTorch, scikit-learn, Hugging Face).
- 3+ years shipping LLM or GenAI applications (RAG, chat/assistants, copilots, content generation, summarization, code or workflow automation).
- Demonstrated experience with agent frameworks (e.g., LangGraph, Semantic Kernel, LangChain, AutoGen, CrewAI, OpenAI Assistants) and tool integration.
- Practical experience with at least one cloud GenAI stack (e.g., Azure OpenAI / Azure AI Studio, AWS Bedrock, Google Vertex AI) and vector search.
- Familiarity with evaluation and guardrails: safety policies, jailbreak/prompt-injection mitigation, PII redaction, and compliance-minded design.
- Strong collaborator with product mindset—able to translate ambiguous problems into measurable outcomes and deliver iteratively.
Preferred Experiences
- Observability & analysis: LangSmith, W&B, Arize, OpenTelemetry; online evaluation & canary strategies.
- Serving & scaling: vLLM, TGI, Triton; Ray/Dask; GPU scheduling; streaming with SSE/gRPC; semantic caching.
- Knowledge & graphs: Knowledge graphs, entity linking, and graph-RAG; schema alignment with enterprise data models.
- Orchestration & workflows: Airflow/Kubeflow/Argo; event-driven agents (Kafka), long-running DAGs with compensations.
- Frontend for AI apps: Streamlit, Gradio, or lightweight web apps; telemetry instrumentation for UX insights.
- Security: Threat modeling for GenAI, data exfiltration prevention, dependency/model supply-chain hygiene.
- Reinforcement learning & multi-agent: RLHF/RLAIF, self-play/simulation for decision-making and policy improvement.
- Containerization & platform: Docker/Kubernetes, IaC (Terraform/Bicep), secrets management, and policy-as-code (OPA).
Nice-to-haves (role-dependent)
- Experience building copilots or enterprise assistants integrated with business systems (e.g., ticketing, CRM/ERP, data lakes, search).
- Familiarity with routing/mixture-of-models and cost-aware policy design.
- Dashboarding/BI (Power BI, Superset) for operational and product analytics.
- Prior work in multilingual, compliance-heavy, or safety-critical domains.
Life at CGI:
It is rooted in ownership, teamwork, respect and belonging. Here, you’ll reach your full potential because…
You are invited to be an owner from day 1 as we work together to bring our Dream to life. That’s why we call ourselves CGI Partners rather than employees. We benefit from our collective success and actively shape our company’s strategy and direction
Your work creates value. You’ll develop innovative solutions and build relationships with teammates and clients while accessing global capabilities to scale your ideas, embrace new opportunities, and benefit from expansive industry and technology expertise
You’ll shape your career by joining a company built to grow and last. You’ll be supported by leaders who care about your health and well-being and provide you with opportunities to deepen your skills and broaden your horizons
Come join our team, one of the largest IT and business consulting services firms in the world
Your future duties and responsibilities
Required Qualifications To Be Successful In This Role
Together, as owners, let’s turn meaningful insights into action.
Life at CGI is rooted in ownership, teamwork, respect and belonging. Here, you’ll reach your full potential because…
You are invited to be an owner from day 1 as we work together to bring our Dream to life. That’s why we call ourselves CGI Partners rather than employees. We benefit from our collective success and actively shape our company’s strategy and direction.
Your work creates value. You’ll develop innovative solutions and build relationships with teammates and clients while accessing global capabilities to scale your ideas, embrace new opportunities, and benefit from expansive industry and technology expertise.
You’ll shape your career by joining a company built to grow and last. You’ll be supported by leaders who care about your health and well-being and provide you with opportunities to deepen your skills and broaden your horizons.
Come join our team—one of the largest IT and business consulting services firms in the world.
Click on Apply to know more.