Silver Touch Technologies is pivoting from IT services to an AI-native platform company. It won a project with RITES Limited to deploy its AI platform, PARAKH, for evaluating infrastructure project reports. This marks a strategic shift and potential entry into a large market.
Silver Touch Technologies Bets Big on AI with RITES Contract
Silver Touch Technologies is transforming into an AI-native platform company, moving away from traditional IT services. The company has secured a significant project with RITES Limited, a government enterprise, to deploy its AI-based Detailed Project Report (DPR) appraisal platform named 'PARAKH'.
What just happened
Silver Touch Technologies announced a strategic shift towards becoming an AI-native platform company. Their first major initiative under this new strategy is the deployment of the 'PARAKH' platform for RITES Limited. This AI platform is designed for the appraisal and intelligence of Detailed Project Reports (DPRs) in the infrastructure sector.
Why this matters
This move signifies a significant change in Silver Touch's business model, aiming for higher-margin, specialized AI solutions. The PARAKH platform's success with RITES could position the company to tap into India's vast ₹111 lakh crore infrastructure market. The deployment of advanced AI, including Large Language Models and a RAG framework, showcases the company's technological capabilities.
The backstory
Silver Touch Technologies has historically been an IT services provider. This transition to an AI-native platform company is a deliberate strategy to leverage advanced technologies like AI and machine learning for specialized applications. The PARAKH platform is described as a production-grade, mission-critical system, not a pilot.
What changes now
The company will focus on developing and implementing AI-native platforms. The PARAKH platform utilizes three self-hosted Large Language Models (LLMs) and over 500 codified engineering rules. It processes over 1,000 historical DPRs using a Retrieval-Augmented Generation (RAG) framework, with a four-layer hallucination prevention system.
Risks to watch
The complexity of executing a mission-critical project for national infrastructure poses a key watch point. Successful delivery is crucial for the company's reputation and securing future government contracts.
Peer comparison
While specific peers in AI-native platform development for infrastructure appraisal are not detailed in the filing, the company competes via the Government e-Marketplace (GeM) portal through Quality and Cost-Based Selection (QCBS) evaluations.
Context metrics (time-bound)
The PARAKH platform uses 3 deployed Large Language Models and processes 1,000+ DPRs within its RAG knowledge base. It incorporates over 500 engineering validation rules.
What to track next
Investors should monitor the successful deployment of the PARAKH platform, its replication across other infrastructure sectors (railways, highways, ports), and further order wins in the government AI space.