The story of Indonesia’s tech sector in 2026 isn’t the funding environment, which has been complicated. It isn’t the IPO pipeline, which has been thin. The story that actually matters for the medium-term trajectory of the digital economy is talent — specifically, AI and machine learning engineering talent — and how short the supply is relative to the demand from Indonesian companies and from foreign employers competing for the same people.

I’ve been talking to founders, CTOs, and university faculty about this over recent months. The shortage is more acute than the topline numbers suggest, and the structural reasons aren’t going to resolve quickly.

The demand picture

Hiring for AI and ML roles at Indonesian companies has accelerated through 2024-25-26 across several segments.

The major digital banks and fintechs (BCA Digital, BTPN, Bank Jago, GoTo Financial, OVO, and others) have invested heavily in AI-augmented underwriting, fraud detection, and customer service. The e-commerce and super-app platforms (GoTo, Bukalapak’s surviving operations, Lazada Indonesia, Shopee Indonesia, Blibli) have built out AI capability in personalisation, logistics optimisation, and content moderation. The traditional banking sector (Bank Mandiri, BCA, BRI) has been catching up after a slow start, and the pace of their AI hiring has stepped up materially.

Beyond the obvious tech and financial sector hirers, there’s substantial demand from telecoms, healthcare, government agencies, and the foreign-owned multinational sector with regional operations in Jakarta and Surabaya. The Indonesian AI talent market in 2026 has more buyers than sellers.

Where the supply is coming from

A few sources. Each has limits.

Domestic universities. ITB, UI, UGM, ITS, and a small number of others produce AI and ML graduates each year. The output is growing but slowly. The quality is variable — the strongest graduates from the top programmes are competitive internationally; the broader cohort is less specialised than the role requirements assume.

Bootcamps and intensive programmes. Several Indonesian and regional bootcamps have built reasonable pipelines into junior data engineering and analytics roles. The output is useful for general data work but doesn’t typically produce the deeper ML engineering talent that companies are actually trying to hire.

Returning diaspora. Indonesian engineers who studied or worked abroad and are returning home account for a meaningful share of the senior AI talent in the market. This pipeline has been important but it’s relatively small in absolute numbers and the people coming back have significant choice about where they work.

Cross-region recruitment. Some hiring happens from Malaysia, Singapore, the Philippines, and Vietnam. The economics are constrained by salary expectations that often exceed Indonesian comparables, and the immigration pathway isn’t always straightforward.

On-the-job conversion. Existing software engineers being upskilled into AI work. This is the largest source of incremental supply but produces engineers with varying depth.

Where the supply is going

The interesting story is that a meaningful share of Indonesia’s best AI talent doesn’t stay in Indonesia, and a meaningful share of what’s left doesn’t go to Indonesian companies.

The pull from Singapore-based regional offices of global tech companies, banks, and consultancies is constant. Compensation differentials are substantial. The work is often more interesting because the projects are larger. The career trajectory is more legible. For a talented Indonesian AI engineer, “work for a global company’s Singapore office while based in Jakarta or remote” is an attractive option that domestic employers struggle to compete with.

Foreign consultancies and global tech firms operating in Indonesia also hire from the same pool. The Indonesian operations of major international consulting firms have built out AI practices that compete directly with domestic employers for talent. Several Australian and US AI consultancies have established Indonesian presences, including the Team400 AI consultancy and various others doing implementation work for both multinational and domestic clients. The competition for talent is one consequence of the broader regional capability building.

The result is a real squeeze on Indonesian companies trying to build AI capability with domestic teams. Compensation has risen sharply — senior AI engineers in Jakarta now command compensation that would have been unusual two years ago — and the retention battle is constant.

What companies are doing about it

A few patterns are visible at companies that have built AI capability despite the tight market.

Investing in internal training pipelines. Several large Indonesian companies have built serious internal AI capability programmes — multi-month immersions, partnerships with universities, dedicated mentor structures. The companies that started this two or three years ago are now ahead. The companies starting now are paying the cost without yet seeing the benefit.

Specialist consulting partnerships for capability transfer. Engaging specialist firms not just for project delivery but explicitly for skill transfer to internal teams. This is the pattern that works when it’s structured as capability building rather than as an outsource. The engagements that work have clear knowledge transfer deliverables and internal staff who actually shadow the consultants.

Strategic compensation positioning. Indonesian companies that have decided to pay at or near international comparables for their AI roles are getting better outcomes than those trying to hold to traditional domestic compensation bands. The economics work out, generally, because the alternative is being unable to staff the projects at all.

Geographic flexibility. Several Indonesian companies have built distributed teams that include Indonesian engineers based outside Jakarta — in Bandung, Yogyakarta, Surabaya, and increasingly remote arrangements. This expands the talent pool meaningfully relative to a Jakarta-only hiring posture.

Acceptance of slower build. The realistic Indonesian AI capability strategy in 2026 is one that builds slowly, retains carefully, and accepts that some ambitions will take longer than the planning slides suggest.

The Indonesian Ministry of Communication and Informatics has been talking about workforce development in digital skills for some time. The national programmes are moving but the scale is small relative to the demand.

What the next 12 months might look like

A few things to watch.

University output growing. New AI-focused programmes at major Indonesian universities are starting to produce graduates in larger numbers. The quality and depth will take time to mature but the trajectory is positive.

Diaspora dynamics. The pace of returning diaspora depends on the relative attractiveness of Indonesian opportunities versus US/Singapore/UK alternatives. The current US visa environment and the changing economic picture in some host countries have shifted this calculus. More high-quality returning engineers are arriving than was the case three years ago.

Compensation continuing to climb. Senior AI engineer compensation in Jakarta will probably keep tracking up. This is partly catch-up to regional levels and partly genuine scarcity. Either way it changes the cost structure of building AI capability domestically.

Foreign company hiring evolving. Whether the Singapore-based regional employers continue hiring Indonesian talent aggressively depends on macro factors and on how the regional tech sector navigates the next 12 months. A pause in this hiring would significantly ease pressure on domestic employers.

The Jakarta Post and Tempo have both run useful pieces on workforce dynamics in the tech sector over recent months.

The honest summary

Indonesia’s AI talent shortage is real, structural, and not going to resolve quickly. The companies that have built capability successfully have invested in internal training, partnered strategically with external specialists, paid competitively, and accepted that the build takes time. The companies hoping to solve this through aggressive recruitment of finished engineers are mostly disappointed — there aren’t enough finished engineers, and the ones that exist have many options.

For the broader Indonesian digital economy, the talent constraint is one of the binding limits on how fast AI capability will spread. The funding cycle matters less than the talent pipeline. The companies and the country that invest most seriously in growing the talent base over the next three to five years will be the ones that benefit most when the constraint eases.